The Homeric Versio Latina

ROBIN SOWERBY

This study examines the versio Latina of Homer's Iliad, first made in the

1360s, from its initial printing in 1537 through all its subsequent revisions up to the end of the seventeenth century.' All complete Graeco-Latin editions of the Iliad available in British libraries have been examined. A stemma listing these editions and indicating the editor or reviser of the versio in each case can be found below (page 189). After its initial printing, the versio was revised eight times in the sixteenth and seventeenth centuries. Conclusions about the process of revision are drawn from a collation of the first book of the Iliad; a sample comprising the first sixteen lines of the versio in its various revisions is also given in an appendix. Many of the examples in the discussion that follows are taken from this sample; the rest are drawn from various parts of Book 1. While the focus is on the Iliad, a historian of the versio Latina cannot ignore the Odyssey, to which frequent reference is made in the early stages of this study. However, no systematic collation of the versio in the Odyssey lies behind anything said of it here. Conclusions drawn from a collation of only one book of the Iliad may be thought to have only a limited validity. There is no reason to suppose that editors were consistent in their practice throughout twenty-four books. On the one hand is the tendency on the part of a less than diligent editor not to sustain the effort put in at the beginning, on the other is the capacity of the diligent and conscientious to improve with practice in performance of the task. In the light of the strange history narrated below, it is to be expected that a collation of all twenty-four books, should anyone think it worth the time and energy to make one, would reveal a number of unpredictable surprises. Nevertheless, limited though it is, this collation of

Book 1, together with discussion of the versio in prefaces, has much of value to tell us of an important and practical aspect of the classical tradition.

' I wish to thank Professor G. N. Knauer of the University of Pennsylvania, who is currently engaged in compiling the entry on Homer for the Catalogus Translationum et

Commentariorum , for his generosity in sharing his findings with me and allowing me with his editor's permission to make use of them here. The article is more accurate than it otherwise would have been. Any remaining inaccuracies are entirely my responsibility. .

1 62 Illinois Classical Studies 21(1 996) as well as something of the nature of editing in the sixteenth and seventeenth centuries. Given the subordinate position of the Greek language in the Latin- based culture of the Renaissance, the literal Latin versions printed e regione opposite the Greek text in Renaissance editions from the mid-sixteenth century onwards were indispensable aids in the dissemination of Greek throughout Europe, offering a close literal key to the original for those whose Greek was poor or who were learning the language in schools and universities. The Latin version, the medium by which most readers approached a Greek classic, had a practical bearing upon its reception in the most basic sense. The Latin version of Homer in the sixteenth and seventeenth centuries has a surprisingly chequered history, noted as such by a scholar of the time, Meric Casaubon, who examined it in the 1650s and wrote at length about it with suggestions of his own for subsequent improvement. ^ This history, which must have both reflected and influenced the reception and transmission of the poems, has not been fully told. It is chiefly interesting today for what it can tell us of what Casaubon calls the iniquitous fate of Homer in his times. Any study of the origin and genesis of Latin versions, however, must go back in time before the invention of printing, and in the present case a natural starting point is the moment when the Greek poet was first translated in the West in the early 1360s by the Calabrian monk Leontius Pilatus for Boccaccio, who then provided a copy for his friend Petrarch.^ In its first manifestation, the version was interlineal, written above the Greek in such a way as to be a verbal key or clavis.'^ Leontius doubtless used his autograph copy to explicate the text when he gave his public lectures on Homer in Florence.

Agostino Pertusi has published the first 147 lines of the opening book of the Iliad from a manuscript copy that includes earlier readings.^ These may be compared with the version in Petrarch's copy of the first book by Leontius previously published by Attilio Hortis.^ In the early stage of his

^ De nupera Homeri editione Lugduno-Batavica , Hackiana . . . item , super loco Homerico

. . . binae dissertationes, auctore Merico Casaubono (John Shirley: Londini 1659). ^ The story has often been told how Petrarch yearned to read the Greek text of Homer that he had been given by an envoy from the Eastern Church, how he tried to learn the elements of the language from another envoy, one Barlaam, who was the teacher of Leontius Pilatus, and how finally he read Homer in his old age through the medium of Leontius' version. See P. de

Nolhac, Petrarque et V Humanisme , 2nd edn. (Paris 1907); J. B. Ross, "On the Early History of Leontius' Translation of Homer," CP 22 (1927) 341-55; A. Pertusi, Leonzio Pilato fra Petrarca e Boccaccio (Rome 1964). "* Pertusi (previous note) 169-82 gives a sample of the Greek with the Latin from the Odyssey ^ Pertusi (above, note 3) 205-19 and 200-04 for his account of the status of this manuscript, which is codex Latinus 7881 in the Bibliotheque Nationale, Paris.

* A. Hortis, Studi sulle Opere Latine del Boccaccio I (Trieste 1897) 543-76, "Appendice in." The manuscript copied for and annotated by Petrarch is codex Latinus 788(3 in the Bibliotheque Nationale, Paris. Robin Sowerby 163 labours we can see Leontius really struggling with the meaning of Homer's words. When Agamemnon angrily dismisses the priest Chryses, Kpaxepov 6' £711 |i\)0ov exe^^ev (25),^ "he laid into him with harsh words," Leontius has contumacem autem ad sermonem inpellebatur precipiendo, where amongst other things he seems not to have understood the tmesis. Confusion about the meaning of the Greek has produced a Latin version that does not make sense. In the version annotated by Petrarch this is changed to contumaci autem sermoni precipiebat. In the next line Kixeico (26), "I find," is first translated as subponam and later corrected to inveniam. When Apollo is invoked as the god oq Xpijoriv d|i{piPePr|Kaq (37), "who have guarded Chryse," Leontius' translation is qui Chrisem proposuisti, where the island is perhaps confused with the priest Chryses. This is later changed to qui Crisem a pueritia defendis, where "from childhood" may be an attempt to account for the tense of the Greek and is perhaps incorporated from a gloss in Leontius' manuscript. When he has recommended the Greeks to restore Chryseis, the priest sums up, xoxe Kev fiiv lX,aaod|ievoi 7teKi9oi|iev (100), "then having appeased him we may persuade him (to stop the plague)." Leontius has tunc forte ipsum rogantes clinabitis, which inadequately renders the participle and produces a transliteration of K>.{veiv, "to make to bend." This later becomes tunc autem ipsum deprecantes mitigabitis, which is nearer in meaning, though the person of the verb is still wrong. Many mistakes are not corrected in the manuscript annotated by Petrarch: xapievx' (39), "pleasing," is not sacrum; xiaeiav (42), "may they pay," is not honorant. When Achilles assures Chryses that no one will harm him as long as he is alive, Kai eni xOovl 5epKO|ievoio (88), "and looking upon the earth," Leontius, who may not have appreciated the middle form, has first in terra viso then in terra conspecto. New confusions are introduced in the later version. Chryses brings ransom gifts that are (XTtepeiaia (13), "numberless." This is rendered by the word miserabilia, though this is perhaps a mistranscription for mirabilia. When Agamemnon agrees to restore Chryseis, Po{)A,on' eyo) ?ia6v aoov e|inevai r\ ocTio^-eoGai (117), "I wish the people to remain safe rather than perish," the earlier version, which makes fair sense {volo ego magis populum sanum esse quam destrui), is completely garbled to become volo ego populum salvum esse 8' quoddestrui. In the earlier version xfi 5eKdxTi dyopTivSe Ka?ieaaaxo laov 'kxxkXzvq (54), "on the tenth Achilles called the people to council," is rendered decima autem ad congregacionemque vocavit populum Achilles, where Leontius shows an often repeated confusion about the use of the enclitic -que in Latin. This inexplicably becomes in nona autem ad agregatorem vocavit populum Achilles , where the number is wrong and a barbarous word has been introduced.

' Quotations of Homer are from T. W. Allen (ed.), Homeri llias (Oxford 1931). The Renaissance vulgate does not differ significantly in any of the passages quoted and discussed. . .

1 64 Illinois Classical Studies 21(1 996)

Quite apart from stumblings over the Greek, there is some incorrect, unidiomatic and peculiar Latin. Alacre (39) should be alacriter. There are many non-classical words like clinabitis and agregatorem above. The Homeric word ^"uvriia (124), "common property," is rendered by the word

ensenia , a non-classical variation from xenium , itself a transliteration of the

Greek ^eviov. When it first occurs (ppeveq (103), "mind" or "feelings," is rendered by essenciae, a rare word not used in the plural in classical Latin. In the proem, 8iaaTTiTriv epiaavxe (6), "they were divided in conflict," is inadequately rendered by diversimode litigaverunt, where the adverb, if

such it is, is a peculiar non-classical formation. Many perfectly good classical words are not always felicitous in the context in which they are being used, as in the case of the following verbs: corrumpebantur autem populi (10) translating 6A,eKovTo 6e A,aoi, "the people perished"; perambulavit (50) for eTctpxeto (of Apollo going against mules and dogs);

ambulaverunt tela del (53) for (iix^'^o (of arrows); ne me instiges (32) ill

translates \\x\ |x' epeOi^e, "don't provoke me." When Agamemnon envisages Chryseis partaking of his bed, ejiov A,exo(; dvciowoav (31), the translation of the difficult word dvxiowaav, occuring only here in this sense

and glossed in the scholia as ^eTexo\)oav , "sharing," is perplexing to say the least: meum tectum contra respicientem The rendering of such ana^ X-eyoiieva and distinctive Homeric vocabulary like the traditional epithets reveals the limitations of the resources at Leontius' disposal.^ He was forced to guess the meaning of many unfamiliar words: d|i(pripe(pea (45), "close-covered," the epithet of a quiver, becomes amplumque pharetram; ix^TtzxiKic^ (51), "bitter" as explained in the scholia or "sharp" as favoured by some modern interpreters, as applied to an arrow, is rendered de bono peuco with a marginal gloss explaining that the arrow was made from cypress. Leontius seems to have created his own Latin word for the occasion. Athene's epithet y^ia-UKCOTiK; (206), "blue-grey eyed," becomes magna oculos Athene He is no more successful in the case of e^iKcoTiiSa (98), "with glancing eyes," which is rendered magnos oculos habentem. Most revealing is the rendering of aiyioxoio Aioq (202 and 222), "Zeus who bears the aegis." Ignorant of the aegis, he was nevertheless prompted by the first part of the epithet to recall the story of Jove being nursed by goats on Cretan Ida: capram lactantis lovis, "goat-milking Jove," turns the sublime into the ridiculous. Before the translation had been begun, Petrarch, who believed that translations should be works of literature, wrote a cautionary letter to Boccaccio advising against an ad verbum version in prose. ^ He quoted the words of Jerome, which he was later to write at the head of his copy of the finished result: si cut non videtur linguae gratiam interpretatione mutari,

Epist. Var. 25, in G. Fracassetti (ed.), F. Petrarcae Epistolae (Florence 1859-63) II 370. ,

Robin Sowerby 165

Homerum ad verbum exprimat in latinum: plus aliquid dicam: eundem in sua lingua prosae verbis interpretetur: videbit ordinem ridiculum, etpoetam eloquentissimum vix loquentem ("If anyone thinks that the beauty of language is not changed in literal translation, let him turn Homer word for word into Latin; let me say something more, let him explicate the same poet with the words of prose, he will see a word-order that is ridiculous and a most eloquent poet scarce able to speak at all"). How perfectly the version '^ of Leontius bears out the verdict of Jerome and Petrarch's worst fears. Notissima nunc grecorum ignorantia, Petrarch had written in his De Ignorantia}^ The Calabrian monk Leontius represented all that Petrarch the humanist wished to turn his back upon. Nevertheless in the same letter to Boccaccio he went on to welcome the translation however it might come, saying that his hunger for noble literature was so great that, just as one who is desperately hungry does not need exquisite cooking, so he expects that food for his soul can be found in whatever form the translation of the great work comes. Given the difficulty of Homer and the prevailing ignorance of Greek in the West, the translation met a real need among the early '^ humanists and it seems to have been copied a number of times. It is remarkable that, although the version was much despised for its general barbarity, it was a hundred years before there is evidence of a better one in circulation, used, it seems, by Andronicus Callistus when he explicated ^^ Homer in the Florentine Academy during his sojourn there from 1471-75. The evidence for this is presented by Ida Maier in her study of the Italian humanist Angelo Politian. She gives a page extract from a study- book comprising notes taken from the lessons of Callistus by one of his pupils, who names his teacher twice subsequently in the notebook. The

'" In his extensive and scholarly study of the translations of Leontius, Pertusi (above, note 3) has become rather partial to his subject. His charitable verdict (442) that the versio of Leontius was not a bad first effort and that it is possible to see its influence in the latest redaction of the versio in the nineteenth century is true in the limited sense that the versio remained ad verbum but it not only ignores the complicated history of successive attempts to revise the versio but also glosses over all the evidence he himself provides of the sheer frightfulness of Leontius' work, its barbaries, from which scholars and humanists recoiled alike, whatever their theory of translation. His list of Latin translations of Homer beginning on page 522 contains a number of inaccuracies. " L. M. Capelli (ed.), De sui ipsius et multorum ignorantia (Paris 1906) 65. '2 See Pertusi (above, note 3) 147 ff., 238 (for the Iliad) and 152 ff. and 200 (for the Odyssey) and P. O. Kristeller, Iter Italicum: A Finding List of Uncatalogued Humanistic Manuscripts of the Renaissance in Italian and other Libraries (London and Leiden 1963-93) III 479a {Iliad, Berlin) and 706a {Iliad and Odyssey, Stuttgart), IV 402a {Odyssey, Komik,

Poland) and 592a ( Odyssey, Madrid). '^ There is an Italian manuscript containing an ad verbum version of Books 1-12 of the Iliad and dated 1410 in the Bodleian library. Though better than Leontius, the understanding of Homer it represents is much more limited than that in the version used by Andronicus Callistus. See R. Weiss, "An Unknown Fifteenth-Century Version of the Iliad," The Bodleian Quarterly Record 7 (1934) 464. There is also a retractatio of a few of the early books of the Iliad by Pier Candido Decembrio made between 1439 and 1441; it remains ad verbum, see C. Fabiano, "Pier Candido Decembrio traduttore d'Omero," Aevum 23 (1949) 36-51. 1 66 Illinois Classical Studies 21(1 996) explication of the Iliad begins with eight lines of a versio Latina followed by notes on individual Greek words.

Iratn cane pelide achillis Funestam que innumeros achivis dolores fecit Multasque praestantes animas orco misit Heroum, ipsosque escam fecit canibus Avibusque omnibus, lovis autem perficiebatur consilio. Ex quo primum separati fuerunt litigantes Atridesque rex hominum et divus achilles Quisque ipsos deorum contentione commovit pugnare '"*

These eight lines are identical to the opening lines of the Latin version of the anonymous Vatican manuscript copied by Bartolomeo Sanvito (the Greek text is copied by John of Rhodes and bears the date 1477) •^ except in two respects. The Vatican version has ipsos vero escam (a minor improvement) and perficiebatur consilium (consilio has no construction and is evidently a mistake on the pupil's part). No other version consulted for this study has funestam for oi)A,o)ievriv, escam (a very loose translation of e^wpia) or praestantes for i(p6{|io\)(;. These are two of four manuscripts containing this version, none of which is complete and of which only the Vatican manuscript bears a date.'^ The study-book seems to be the earliest, but even though it cannot be said that Andronicus is the reviser (it may have been composed much earlier), its use by this notable Byzantine emigre in the Florentine Academy suggests—what indeed proves to be the case—that it represents a marked advance in Homeric understanding and scholarship beyond anything in the cultural substratum of southern Italy which had

'"^ produced Leontius. Indeed it probably represents a new start. These are the renderings in the version used by Andronicus of the words and lines

'** I. Maier, Ange Politien: La formation d' un poete humaniste (1469-1480) (Geneva 1966) 58-59, "Appendice VI." (See also 30-31 for further comment on Callistus.) The manuscript is codex Laur. 66. 31, Florence (Kristeller [above, note 12] V lib). In Maier' s transcription there are three mistakes: pelidae for pelide, Multaque for Multasque and proficiebatur for perficiebatur. In this last word the prefix is contracted; the same contraction occurs three times elsewhere on the same page, where it is correctly transcribed as permanens, per geminum and perdo. For further discussion of this codex, see G. Resta, "Andronico Callisto, Bartolomeo Fonzio e la prima traduzione umanistica di Apollonio Rodio," in Studi in onore di Anthos Ardizzoni (Rome 1978) 1094 ff. '^ Cod. Vat. gr. 1626. See Pertusi (above, note 3) 139 n. 2, who first drew attention to this version and cites the opening lines, and Kristeller (above, note 12) II 389a and VI 318a. The opening pages of this Greek and Latin Iliad, beautifully illustrated, are reproduced in Rome Reborn, ed. by A. Grafton (Washington, London and New Haven 1993) 7. '^ The other two are: MS A. 1414, Biblioteca Comunale d'Archiginnasio, Bologna (believed to be an early 16th c. copy of the Vatican codex; see Pertusi [above, note 3] 139 n. 2 and

Kristeller [above, note 12] V 1 lb) and MS V. a. 19, Kungliga Biblioteket, Stockholm (believed to be late 15th c; Kristeller [above, note 12] V 1 lb). I am indebted to Professor Knauer for communicating to me his preliminary conclusions about this group of manuscripts. '^ If evidence of this is required, then it can be seen in the translation of particles (often an easy indication of indebtedness): Where Leontius has autem for 6e and vel for r\, in the Vatican manuscript they are vero and aut. Robin Sowerby 167 quoted from Leontius in the order in which they occur above: aspemm vero verbum dicebat (25), reperiamur (26), gubernas (37), tunc ipsum propositum reddentes suadebimus (100), gratum (39), puniantur (42), in terra existente (88), innumerabilia (13), volo ego populum salvum esse non perire (1 17), in decimo vero ad contionem vocavit populum Achilles (54), communia (124), mens (103), separati fuerant litigantes (6), peribant autem populi (10), adivit (50), ibant (53), sed valde non me stimula (32), meum ledum ministrantem (31), ab utraque parte co-opertam (45), amarum (51), dea caesos oculos habens Minerva (206), nigros oculos habentem puellam (98). The author, who has a sure command of Latin, also made consistent use of the glosses he had at his disposal contained in the Greek scholia, which might have saved Leontius from some of his misunderstandings. The scholiast glosses aiyioxoio (202) as aiY{8a exovxoq and explains that the aegis is the fiery shield of Zeus. In the Vatican version this is rendered at 202 as aegidem tenentis lovis and at 222 as tenentis clipeum lovis. A good Byzantine scholar would not have needed to consult a gloss to render the aegis of Zeus correctly, but the scholia were useful for more difficult words. For instance, the rendering of diKpripecpea (45) as ab utraque parte co- opertam derives from the gloss xfiv eKatepoGev eaKeTtaoiievov and the rendering of exeTiEDKec; (51) as amarum derives from the gloss exov

TciKpiav . The note in the scholia on the epithet of Apollo, I|iiv0£ij (39), explaining that it is derived from a word for mice because Apollo had intervened to save one of his temples from a plague of them, is responsible for the rendering Murum interfector. Occasionally reading of the scholia leads to pedantry: no8dpKr|(; (121) is rendered pedibus sujficientem habens (rather an understatement for one who, like Achilles, excelled in running), which is probably prompted by the Greek ejiapKeiv xoic, Tioai S-uvd^ievoq.

In the light of this care, it is difficult to account for the weak rendering escam for eXwpia (4), which is glossed as £XKx>(5\iaxa and (5iiapay\iaxa. Even more surprising is the definite mistake at 29, where Agamemnon says he will not release Chryseis before old age (ynpaq) comes to her in Argos. In the Latin we have honor, which is a version of yepaq, perhaps to be explained by a garbled reading in the Greek manuscript from which the translator was working, as, after Leontius, the version as a whole is

strikingly error-free. It is also composed for the most part of Latin words that have a good classical pedigree and are idiomatic in context. Although

it is far superior to that of Leontius and would have provided a much better base for subsequent improvement, it seems to have remained unnoticed.

There are no traces of it in the printed version. The first printed Latin version of the Odyssey, Andrea Divo Justinopolitano interprete, ad verbum translata, published in Venice 168 Illinois Classical Studies 21 (1996) without the Greek text in 1537,'^ closely follows the Latin of Leontius Pilatus. '^

Leontius Virum michi pande, Musa, multimodum, qui valde multum Divus Virum mihi die musa multiscium, qui valde multum

Leontius Erravit, ex quo Troie sacrum civitatem depredatus fuit; Divus Erravit, ex quo Troyae urbem depopulatus est;

Leontius Multorum hominum vidit urbes, et intellectum novit: Divus Multorum autem virorum videt urbes, et mentem cognovit;

Leontius Multas autem hie in ponto passus fuit angustias proprio in animo Divus Multas autem hie in mari passus est dolores, suo in animo

Leontius Redimens propriam animam et reditum soeiorum. Divus Liberans propriamque animam et reditum soeiorum.

Leontius Sed non sie socios salvavit, desiderans lieet; Divus Sed neque sie soeios liberavit, eupiens quamvis:

Leontius Ipsorum enim propriis stultitiis perierunt; Divus Ipsorum enim propriis stultitiis perierunt;

Leontius Stolidi qui per boves yperionis solis Divus Fatui qui boves Hyperionidae Solis

Leontius Comederunt, nam hie istis abstulit reditus diem. Divus Comederunt: sed hie his abstulit reditus diem.

A note to the reader makes it clear that the version, though printed independently, was intended to be used with the Aldine Homer, to whose pages it corresponded (sig. a 2^)-?^

Quam in excudenda Homeri Iliade latina diligentiam antea adhibuimus, eandem nunc etiam praestitimus in imprimenda Odyssea, itaque quod ad singulas chartas earumque numeros, paginas et versus attinet, eadem hie omnia sunt, atque in Homero graeco ex aeditione Aldina, versuum numero insuper a nobis adjecto. Quod ideo a nobis factum est, ut qui Poetam hunc graecum legunt, eundem etiam habeant latinum ita excusum, ut uno intuitu omnia in utroque videre possint.

In the Iliad, however, at least in Book 1, Divus' version represents virtually a new start. When Leontius began translating Homer, the task was almost beyond his capabilities; however, by the time he had reached the Odyssey his understanding of Homeric Greek and his command of Latin had

'^ Homeri . . . Odyssea, Andrea Divo Justinopolitano interpreter ad verbum translata (Jacob aBurgofrancho: Venetiis 1537). '^ For the opening lines of Leontius' Odyssey, see Pertusi (above, note 3) 441. ^° There are three Aldine editions of Homer: 'Onripou 'Vkyac, etc., Homeri Ilias etc., ed. by A. P. Manutius, 2 vols, (in aedibus Aldi: Venetiis 1504, 1517 and 1524). The first edition is without pagination, which is added to the subsequent reprints. Robin Sowerby 169 improved considerably. But Divus must have had Leontius' Iliad, for there is too much odd vocabulary common to them both which cannot be explained by a common word-list. Divus generally made sense of Homer, but occasionally follows Leontius' mistranslations, as for example in lines

352 and 564 of Book 1:

|j,fiTep, inei n' excKeq ye Jiivuv0d5i6v Tiep eovxa

Leontius Mater postquam me peperisti parve viteque existentem Divus Mater postquam me peperisti brevis temporis existentem

ei 5' OTJXCO xom' eaxiv, e|iol \ieXXei (piXov eivai

Leontius Si sic hoc est michi debet amicum esse Divus Si autem sic hoc est: mihi debet charum esse

In 352, the sense requires a causal conjunction, not the temporal postquam. In 564, p.e^^ei (glossed eoiKe) is more correctly rendered in the Vatican version by videtur. Divus was helped in his revision by the Attic paraphrase of the Homeric Greek which could be extracted from the scholia minora and had been the basis of the Byzantine paraphrases of Michael Psellos and Manuel Moschopoulos,^' and of the paraphrase made in Renaissance Italy by Theodorus Gaza.^^ This Greek interpretatio had been transmitted with many of the texts of the Iliad in manuscript, and was printed with the text for the first time in the edition of the Iliad and Odyssey published at Basle in 1535.^3 From extant papyri, the scholia minora can be dated to Athenian times. However, not all the material they contain can be of Athenian origin. 2^* The Athenian schoolboy, like the beginner at Rome and Byzantium, might have needed to know that aei6e, the second word of the poem, is a Homeric form of a6e, but the reductive gloss that follows, ^eye, is probably of a later date. Psellos begins, tfiv opynv eine, co 0ed. Moschopoulos begins more pedantically, co 0ed KaA,A,{o7te, eirte ocKpiPox; xx\ eppeX.{a jroiriTiKTi. Gaza begins, tt^v opyriv eiTte fipiv, co 0ed. There are many other reductive interpretations which progressively wander away from Homer's words in an attempt to explicate the poem in the simplest possible

^' Pertusi (above, note 3) 455-57 prints the opening 42 lines from the first book of the Iliad of the paraphrases of Psellos and Moschopoulos. ^^ The paraphrase of Theodorus Gaza is printed in N. Theseus (ed.), 'Onripov 'IXmq iieia

TtaA^iaq napaippdaeoq . . . xot) 0eo5(opo\) Fd^a (Florence 1811-12). ^^ 'OfiTipou 'iXidq Kttl '05\)ooEia netd xfii; z^y\"(r\csxoc,, Homeri llias et Ulyssea cum interpretatione (apud J. Hervagium: Basileae 1535). '^^ See H. Erbse, Scholia Graeca in Homeri Iliadem (Berlin 1969) I xi, where Erbse writes of the scholia minora: "nam reliquias eorum interpretamentorum continent, quae pueri

Athenienses Homeri intelligendi causa inde a quinto a. Chr. n. saeculo in schola discebant. . . Etiam nonnullis papyris satis antiquis prorsus evincitur fundamentum illorum scholiorum ante aetatem Hellenisticam iactum esse." 1 70 Illinois Classical Studies 21(1 996) form. The scholia minora transmit a variety of material of varying value, ranging from the important explanations of rare words and ana^ Xeyoiieva that re-appear in Eustathius and the scholia of Venetus A to the crudest Byzantine gloss. Many of Divus' renderings ultimately derive from a not very discriminating use of these scholia or from the paraphrase based upon them and are therefore at one remove from the primary text. For example, Psellos in his paraphrase consistently has |iaKpop6X,oq for eicriPoAoq, which

Divus translates longejaculans. Longe is the Latin equivalent in the early Graeco-Latin lexica of fiaKpo- in compound words, and Divus has chosen jaculans to render -Po^-oq. It is an unthinking choice, for jaculor is not a

suitable word for an archer, as it denotes the action of throwing a missile from the hand. It is doubtful whether longe can mean "from far off." Literally, longejaculans means "throwing ... a long way off." Leontius' procul sagittans, "shooting arrows from afar," does convey coherent

meaning, whereas the Latin of Divus is stranger than the Greek, and, unlike the Greek, cannot be explained etymologically. The scholiast has eKaGev, 6 eaxi TioppcoGev ^aXkovxoc,, eijaToxou xo^oxo'u. Divus has not heeded the

scholiast but has seen fit to alter Leontius. He may have been working directly from the paraphrase. This seems unlikely as |iaKpoP6^0(; has not found its way into the printed Graeco-Latin lexica. Or he may have picked

up the phrase from another translator.^^ But however it came to him, the result is scarcely felicitous. Another more felicitous word that is probably

derived from the paraphrase is pulchricoma, translating KaA.XiK6|iri glossing

•n{)K0|i0(; (36), the epithet of Leto. Right at the outset of its printing, the history of the v.ersio is a complicated affair. There are two editions of Divus published in Venice in 1537, one by Jacob a Burgofrancho, who also published Divus' translations of Aristophanes in 1538 and of Theocritus in 1539, and another by Melchior Sessa. 2^ Both contain the same general preface by Divus addressed to Pier Paulo Vergerio the younger, and the same claim about the relation to the Aldine edition, except that the edition of Sessa omits versuum numero insuper a nobis adjecto, as, though the pages are numbered, the lines are not. Nor do the page numbers in the edition of Sessa actually correspond, as do those of Jacob a Burgofrancho, to those of the Aldine edition. It might seem from this that the edition of Sessa is simply an inferior pirated

edition, except that it is surprising to find that the version has been revised

and that the revisions, in about eighty places in Book 1, are for the most part improvements on what seems to have been the authorised version. There are some corrections of mistakes. For example, the rendering of the Greek

^^ Professor Knauer has pointed out to me that longe jaculantis translating ekcxtoio occurs in the opening line of the version of the Homeric Hymn to Apollo by Georgius Dartona Cretensis.

See volume two of the edition of Divus published by Burgofrancho, f. 219. ^^ Homeri . . . Ilias, Andrea Divo Justinopolitano interpreter ad verbum translata (Jacob a Burgofrancho: Venetiis 1537) and Homeri Ilias, Andrea Divo Justinopolitano interprete, ad verbum translata (M. Sessa: Venetiis 1537). Robin Sowerby 171

eKTiay^OTaTe (146) is changed from splendissime to formidolissime. When Chryseis is said to be not inferior to Clytemnestra ov bi^aq o\)6e (pvr\v (115), neque corpore neque aetate is changed to neque corpore neque forma (which had been the rendering of Leontius too). When Achilles complains to Agamemnon, d>.A,a xa \xkv no^icov e^87ipd9o|iev, td 585aoTai (125), "But what we took from the cities by storm has been distributed," sed partim quidem ex urbibus depredati sunt, partim vero divisa est is changed to sed quae quidem ex urbibus depredati sumus, haec

vero divisa sunt. At 452 (191 dvdaaei(;), generose imperas is changed to patenter imperas . There are also changes that are the result of reading the scholia and other ancillary material. When Agamemnon says that he is minded to take Chryseis home, he envisages her weaving at the loom and ejiov Xixoc, dvxiocoaav (31); instruentem, which implies that she is to be a

maidservant, is changed to mei lecti participem existentem , which is an ugly formulation but does translate the meaning as clearly given in glosses in the

scholia, dvTi^a|iPavo|j.evriv and iiexexo-uaav . Agamemnon quite clearly prefers her to his wife Clytemnestra, Ko\)pi5{r|(; dA,6xo\) (114), which is changed from puellari to legitimae, a more thoughtful rendering (supported by Liddell and Scott) and one that is not this time prompted by the gloss in the scholia, TuapGeviicfiq. When Achilles tells Thetis of Apollo's anger,

which causes the Greeks to die eTtaaavxepoi i3S3),frequentes is improved to coacervati from the glosses e7idX,A-ri^oi and TtDKvoi. Yet despite these corrections and intelligent renderings of difficult words, the errors of Leontius discussed above (in lines 352 and 564) remain uncorrected. Since it was the "authorised" version of Divus that was reprinted, the improvements in the version published by Sessa did not subsequently become part of the tradition, but they do show how much room for improvement there was in the authorised version of 1537.

The version of Divus was designed to be read with the Greek text. It was only a matter of time before an enterprising printer produced an edition of Homer which contained both text and translation (rather like the modem Loebs). The first Graeco-Latin edition of Homer, published in Basle in 1551,2'' ^25 superintended by the scholar-printer Johannes Oporinus, who as a young man had worked for the great Froben and had received financial support from Erasmus. He had been professor of Greek at the Collegium Sapientiae in Basle before becoming a printer and he had also acted for four years as an assistant to Paracelsus. In the preface to the edition of Homer, he laments the ignorance of Greek in northern Europe, a deficiency he doubtless felt it was his mission to remedy, for he produced many editions of the Greek classics first published by him north of the Alps. Nothing is

Graece , adjecta versione Latina ad verbum, ex diversis doctissimorum virorum translationibus concinnata etc. (perN. Bryling. & B. Calybaeum: Basileae 1551). 1 72 Illinois Classical Studies 21(1 996) known about the cirumstances surrounding the composition of the first printed edition of the versio by Divus, but from a bundle of letters written by Joannes Vuolphius of Zurich,^^ we can see something of the effort that went into the making of what promised to be a splendid landmark in publishing history. Oporinus evidently entrusted the main task of the edition to Vuolphius. The first two letters concern the Greek text. Vuolphius has an Aldine edition and Oporinus supplied other unnamed editions, which Vuolphius proposed to compare with the Aldine.

Subsequently, he decided to reprint the Strasbourg text, first edited by Joannes Lonicerus, which he judged to be better than the Aldine. In the letter of July 1544,^9 the plan for the edition has been clarified and is to include a versio (f. 369):

Ad XV Cal. Jun. dedisti ad me litteras Oporine, ex quibus quod in Homeri versione fieri velles intellexi ... At q[uia] hoc tuum institutum (ut prius) celari debet, nolui multos huius laboris fieri participes, ne hac ratione plures etiam tuum consilium resciscerent, eoque Sebastianum accessi, atque effeci ut is esset laborum socius, quod et tibi probari maxime videbatur. Itaque accepi exemplar Collini, et illud Glareani quod tu misisti ad Frisium, et aliam cuiusdam Itali versionem, quas omnes confero, et quod optimum in his videtur, in banc meam pono.

From this we can conclude that the edition was to be a secret, that Vuolphius had assistance for the versio (in addition to Sebastianus, in the letter of February 1545 [f. 370] he mentions another scholar engaged in its composition by the name of Clauserus) and that Oporinus had gone to some trouble to provide Vuolphius with two existing versions^^ in addition to the Italian manuscript that he had in his possession. Vuolphius says he will collate the three and take the best for his own, a very reasonable plan. In the last letter (f. 372), dated August 1545, he tells Oporinus that he has sent the Iliad translation and promises that Sebastianus, who together with

^^ See M. Steinmann, "Johannes Oporinus," Easier Beitrdge zur Geschichtswissenschaft 105 (1967) 141^2. The letters he mentions and to which reference is made in this study are available in manuscript in the library of the University of Basle (Universitatsbibliothek, Fr. Gr.

I ll,fols. 367-72^). ^^ Steinmann (previous note) 142 gives the date as 1545. The final number is not clearly legible. However, the rest of the letters are in chronological order and the content subsequently moves on to later stages of the production, so that 1544 seems more likely. "^ A version of about a third of the Iliad (ending at 8. 26) by Henricus Glareanus (Heinrich Loriti, 1488-1563) survives in manuscript; see Ljubljana, Narodna in Univerzitetna knjiinica, 197 (Kristeller [above, note 12] V 443b). For the first fifteen lines of the version, see P.

Simoniti, "Glareanov latinski prevod Iliade," ZAnt 34 (1984) 192 f. Having examined the manuscript, Professor Knauer (to whom I am indebted for information about the two versions mentioned in this note) dates it between 1520/21 and 1539. A complete Latin version of the Iliad, bearing the date 1536 and the name Rudolphus Gualtherus Tigurinus, may be a copy of the version of his teacher Collinus (Rudolf Ambuhl, 1499-1578). See Zurich,

Zentralbibliothek, C. 1 19b (Kristeller [above, note 12] V 153b). Robin Sowerby 173

Clauserus had worked on the Iliad, would translate the Odyssey before the year was out. The printing difficulties were considerable, for Oporinus' underlings were at first unable to read the characteres of the manuscript (fols. 370 and 371). The edition did not appear until 1551. In the printer's preface to the reader, Oporinus idealistically explains that he has added the ad verbum Latin version e regione , optimis quibusdam interpretibus usus, having used the best interpreters, so that youngsters so far put off by the difficulty of the task might have a text that they can read by day and by night (echoing Horace's vos exemplaria graeca I nocturna versate diu versate diuma, AP 268-69) and so that older people can get a taste of Homer's wisdom and can gradually with some effort attain a perfect understanding of the language: Non ut vivam, ob hoc, praelectoris vocem

. . . parvi ducendam . . . et mutis tantummodo hisce Uteris, quibus freti, saepissime turn sensus turn ipsorum thematum diversa investigatione decipiuntur, inhaerere studeant ("Not that they may on that account consider the living voice of the teacher to be of no account and may be keen to stick to the dumb texts relying on which they are often deceived in their various enquiry both into the sense and the themes themselves"). As a former teacher of Greek, who may well have explicated the text of Homer in the Collegium Sapientiae in Basle as Callistus had done in Florence, Oporinus knew the limitations of the teach-yourself method. AH the indications are that the edition was a serious undertaking, the purpose of which was not primarily to make money or even to achieve a publishing coup, but to fulfil a real educational need in the best Erasmian tradition. It therefore comes as a great shock to find that the versio is a frightful mess, much inferior to the Venetian Divus in either manifestation. In the first eight lines there are three errors of tense (5, 6 and 8). The version of line 8 is particularly bad:

t{q t' ap aq)(oe Gedjv epi6i ^\)veT|Ke |idxeo0ai;

Quisque sane ipsos Deorum contentionem commiserit pugnare

Whereas in Divus intelligent use of the scholia had sometimes led to improved readings, particularly in the version published by Sessa, the 1551 version is quite unintelligent in its incorporation of glosses. Vuolphius had before him the edition of the scholia minora published in 1539 at Strasbourg with the title '0|ir|pot) 'E^tiyriTfi^, Homeri interpres?^ He confessed (f. 369), multum etiam Graecis scholiis iuvari videor quas Vendelinus Rihelius Argentinae impressit ("I seem to be helped by the scholia which Vendelinus Rihelius has printed at Strasbourg"). Occasionally his use of the scholia bears good fruit. The phrase eminus ferientis for ektiPo^ou (4), though ugly, makes better sense than

^' 'OnTipo\) kfy\yr\zi\c,, Homeri interpres etc., ed. by Jacobus Bedrotus (per Vuendelinum Rihelium: Argentorati 1539). 1 74 Illinois Classical Studies 21(1 996) longejaculantis of Divus, and dvxiowoav (31) is rendered correctly as participantem. But, generally speaking, the glosses incorporated into the 1551 version are of the most reductive kind: die (1) from A,eye for aeiSe; Graecis (2) from "EXXr\(5\ glossing 'Axaioiq; fecit (2) from e7to{r|ae for 6fiKe, which must have a more forceful meaning than "made," if only something like "wrought" or perhaps "ordained" (as in Liddell and Scott); generosas (3) from yevvaiouc; for i99{|xo\)(;, where yevvaio-uq is a second gloss following iaxv pa<; (giving the translation /(9rre5 in Leontius, Divus and subsequent versions). The secondary gloss in this instance is not reductive; it simply adds to the associations of the Homeric word. But it is typical of the 1551 version that it wanders away from the primary meaning; the choice of secondary glosses results in a consistently weakened and watered-down version, which is not actually what a beginner or anyone else for that matter requires. It is confusing for a beginner to be faced with die for aei6e. Conversely, even a beginner does not need to be told that the birds that will rend the corpses of heroes are earnivoris (5), derived from aapK0(pdY0i(;, a most redundant Byzantine gloss. The 1551 version has a persistent, unhelpful and almost perverse tendency to avoid direct translation. Apollo's arrows that are rendered in all other versions by the obvious equivalent sagittae are jacula (51) or missilia (53), not the most appropriate words in context. Equally silly is the Latin chosen for vvkxI eoiKwq (47), nocte adsimilatus , rendered in all other versions as noete similis. Where translationese might be helpfully suggestive in a phrase like 8eivTi 6e KA,a7yr| (49), which in other versions is terribilis autem elangor, in

1551 it is vehemens vero sonitus. At least this is good Latin, but of all the versions that of 1551 is decidedly the worst in its Latin usage: Perdentem (2) for ot)Xonevr|v needs an object. All the versions are replete with cumbersome and ugly phrasing but laeerationes fabrieavit (4) is worse than any of the other renderings of eA-copia xe^xe and does not have the merit of being any more accurate. In line 12, eurrentes for Godq is a strange choice. The general lack of grip may be illustrated in one telling final point. At line

202, aiyioxoio is rendered aegidem habentis , but at 222 is the extraordinary translation eapram laetantis. Evidently the Italian version which Vuolphius had in his possession was by Leontius, not by Divus. Given that Divus had been printed in Paris and Lyons as well as in Venice,^^ it is strange that Oporinus, who provided Vuolphius with recent helpful Greek texts and two manuscript versiones, did not also provide him with the printed version of Divus. In the event, Vuolphius and his assistants were clearly not up to the task. From his address to the reader, it is evident that Oporinus knew the difficulties faced by students who were wholly reliant on a versio without the mediating influence of a praeeeptor . Either he did not give clear instructions to his

Homeri Ilias etc. (in officina C. Wecheli: Parisiis 1538 and Lugduni 1538). Robin Sowerby 175 team in Zurich, who were working at some distance from his own seat of operations in Basle, or the team did not have the wit to think them through and carry them out properly. Their versio is not a proper clavis, as its words are at one remove from the Greek. Johannes Vuolphius is the John Wolf of ^^ Zurich (1527-157 1 ) who later became professor of theology there. At this time he was a young man and the Homer edition seems to have been his first scholarly venture. There is no record of further work done by him on any of the Greek classics. Oporinus had been a professor of Greek; could he not have found an established Greek scholar for a task in which he evidently believed? Most of the Homeric scholarly enterprise in Italy had been undertaken or supported indirectly by the expertise of Greek-speaking representatives of the old Byzantine culture. ^'^ The first Graeco-Latin edition produced by Oporinus was also the first major development in Homeric publishing that did not have this sustaining background presence. The venture sadly bears out his own lament for the ignorance of Greek in northern Europe in his time. Oporinus cannot escape blame for the iniquity of the version. In fact, as its commissioning editor the prime responsibility lay at his door. It is not as if the work was a rushed job either. Six years elapsed between

Vuolphius' submission of the version of the Iliad and its printing in 1551.

Yet, if it is true that Oporinus was responsible for upwards of 700 books, at one time employing more than fifty workmen,^^ then it is hardly surprising that he had little time to scrutinise the quality of the scholarly work he commissioned. When he came to organise a second edition (published in Basle in 1561),^^ Oporinus turned to Sebastianus Castalio, an experienced scholar who immediately saw what Oporinus must also have known after publication if not before that the versio was wholly inadequate. In the case

See Steinmann (above, note 28) 142. The editio princeps (Nerlius: Florenc Lives of Homer attributed to Plutarch and Herodotus and the Oration of Dion, was the work of the Greek Demetrius Chalcondylas, from whom the first publication of all the Greek material relating to Homer may be traced in a direct line. A Greek pupil of his, Janus Lascaris, also a protege of the Greek Cardinal Bessarion, did the work for the editio princeps of the scholia minora, LxoX^ol Tia^aict tmv Tidvu 8ok{hidv ei<; ttiv 'Onripou 'lA,id5a, ed. by J. Lascaris (in gymnasio Mediceo; Rome 1517), and for the editio princeps of the Quaestiones of Porphynus, rioptpupiou (piA.oo6(po\) 'OntipiKot ^TiTriiiaia etc., Porphyrii Philosophi Homericarum Quaestionum Liber etc. (in gymnasio Mediceo: Rome 1518). Lascaris and Marcus Musurus had prevailed upon the Medicean Pope, Leo X, to establish the Greek college, the Gymnasium Collini Montis, on the Quirinal, of which Lascaris became rector and professor of Greek. From this college came the first edition of Eustathius, prepared by two of Lascaris' pupils:

EuoxaBwu . . . riapCKpoXai eiq xr\v '0|ifipou 'IA.id5a Kai 'OSiiooeiav, \ie-za. nivaKOc,, by Matteus Devarius, with the text edited by Nicolaus Maioranus, 4 vols, (in gymnasio Mediceo: Rome 1542-50). ^^ See J. Aldis, The Printed Book : The Original Manual . . . revised . . . by John Carter and Brooke Crutchley, 3rd ed. (Cambridge 1951) 26.

Homeri opera Graecolatina quae quidem nunc extant, omnia, . . . cum Latina interpretatione . . . In haec operam suam contulit S. Castalio, etc. (per N. Brylingerum: Basileae 1561). 1 76 Illinois Classical Studies 21(1 996)

of the Iliad, he declares quite openly that he has substituted without revising the versio of Divus that had recently been reissued also without revision by Crispinus in Geneva in 1560. As Crispinus had not yet produced an Odyssey (this did not materialise until 1567), Castalio was forced to revise the 1551 version of Sebastianus himself. He reported that except for the later books, which were not quite so bad (Sebastianus, like Leontius, evidently improved with practice but neither scholar had the intellectual or professional pride to correct his own work), he was not so much a corrector as the original translator: innumeris in locis non tarn correctorem quam

interpretem . The editions of Divus published in Venice, Paris and Lyons must not have been known in Basle, as Castalio, who admitted that more work needed to be done than he was prepared to do, would not have scrupled to reprint the Latin Odyssey of Divus as well. He implies that he had undertaken the task unwillingly: Effiagitavit a me Oporinus primum ut

in Homerum operant aliquant navarem . . . maluissem in meliora operant impendere. At the end of his preface, Castalio, who had taught Greek but was more of a theologian than a classical scholar, confesses that his interest in Homer had been a youthful enthusiasm before he had fully seen the Christian light. Once again, Oporinus failed to engage a scholar with the

dedication necessary for the task, though perhaps it could be said that Castalio served Homer better than Vuolphius and his assistants. The Basle editions of Homer are magnificent folio volumes, handsome in appearance, even if that appearance to some extent belies their contents. The Genevan editions that followed, emanating from the press of another scholar-printer, Jean Crespin, represent another publishing landmark and also mark a new stage in the history of the versio. Crispinus' first edition of the Greek only in 1559 is an enchiridion,^^ a small sedecimo volume of the kind produced earlier in the sixteenth century for the Latin classics by the Lyonese publisher Sebastian Gryphius. ^^ His preface recalls the famous anecdote in Plutarch that Alexander the Great had carried around an enchiridion of the Iliad on his campaigns in Asia.^^ More ambitiously, in the next year he brought out the first sedecimo Graeco-Latin text: '^° Quod plerique typographi excudendis hac forma Latinis Poetis magna cum laude factitarunV. id nos his primum typis cum magna et operarum et operis dijficultate in gratiam studiose iuventutis, quae o^vSepKla valet, in Graecis

poetis tentavimus . The Latin version e regione (on the left hand side) is for those tyrones who are new to Greek literature. He says further: Quotquot sane antehac edita ad verbum fuerunt exemplaria diligenter contulimus, adhibita insuper veteri et manu scripta Itali cuiusdam interpretatione, quam

^' 'HpcoiKa. 'OfiTipou 'IA,id(;, Homeri Ilias, id est, de rebus ad Troiam gestis (e typographia J. Crispini Atrebatii: [Geneva] 1559). ^^ On Gryphius, see Aldis (above, note 35) 28. 3^ Plutarch, Alex. 8. 2. ^^ With the same title as the 1559 edition (above, note 37). Robin Sowerby 177 nos ex magni Budaei bibliotheca accepimus. Although this gives the impression that he has collated versions which had been previously published (which can only have been those of Divus and the Basle edition of 1551), in fact he has simply reprinted the "authorised" version of Divus, presumably from the manuscript in Bude's library, with no revision. Some years later came the Graeco-Latin edition of the Odyssey (preceded as in the case of the Iliad by a plain text of the Odyssey in the same year, 1567).'*' The preface to the reader by Crispinus strikes a new note. Where previously there had been self-congratulation (in Crispinus' case quite justifiably since a sedecimo Graeco-Latin edition is indeed a great technical feat), here for the first time is self-defence. Just as many early humanists despised the ad verbum method and execrated the version of Leontius Pilatus, a persistent strain in humanism continued to look askance at versiones composed on the ad verbum principle. On their first printing in 1537, the versions of Divus had been immediately criticised by the German humanist and pupil of Melanchthon, Joachim Camerarius, in the preface to his own explication of the first book of the Iliad, published in 1538, to which he appended a translation in Latin hexameters.^^ Even if the translator finds good Latin words, he argues, what is the profit, if they are not made to obey the laws of Latin syntax and grammar? Given differences between the two languages, whoever tri'es to learn Greek by way of an ad verbum version is likely not only to fail to learn Greek but also seriously to corrupt his Latin. The ad verbum versions corrupt both the matter and manner of the original as well as obscuring and degrading them and so should be shunned like the plague. In school Camerarius recommends that they should be used, if they must be used at all, as a warning example by which pupils can learn a proper method of translation. In the 1567 preface, Crispinus apologises to the reader for the several years' delay in producing the Odyssey, saying that it was due to second thoughts about the versio (perhaps prompted by reactions to his Graeco- Latin Iliad similar to that of Camerarius to the same version in 1538). He has taken great pains to see that the version is more correct than in previous editions. He then writes at length about the deficiencies of such versions, referring to them collectively as a horridum et spinosum loquendi genus whose practitioners sometimes contrive to make the Latin more obscure than the Greek (he may have had the Basle edition of 1551 in his sights here). How often they fail to make proper sense. How poorly the Latin words correspond to the Greek. And how ludicrous is the final result. The

'Onfipou '05iL)Ooeia , Homeri Ulyssea, id est, de rebus ab Ulysse gestis (e typographia J. Crispini Atrebatii: [Geneva] 1567).

'* Commentarius Explicationis primi libri Iliados Homeri, loachimi Camerarii . . . Eiusdem libri primi Iliados conversio in Latinos versus, eodem auctore etc. (impressum in officina C. Mylii: Argentorati 1538). 1

1 78 Illinois Classical Studies 21(1 996)

ad verbum method is unsatisfactory because there is so much debate about the actual meaning of the words in so many places; even in the case of a talented interpreter religious adherence to the words of the original produces something that is alien to good linguistic usage and knowledge; what is required in the interpretation of words is something proper that makes sense. He then claims to have provided a versio that, while following the original closely, is neither obscure nor disagreeable, having made use of the work of a scholar who has examined the better versions and made corrections of his own. Let us compare the opening lines of the printed versions of the Odyssey, juxtaposing Divus (1537), Sebastianus (1551), Castalio (1561) and Crispinus (1567):

1537 Virum mihi die musa multiscium qui valde multum 1551 Virum die mihi Musa versutum qui valde multum 1561 Virum die mihi Musa versutum qui valde multum 1567 Virum die mihi Musa versutum qui valde multum

1537 Erravit ex quo Troyae sacram urbem depopulatus est 1551 Erravit postquam Troiae sacrum oppidum evertit 1561 Erravit postquam Troiae sacrum urbem evertit 1567 Erravit postquam Troiae sacrum oppidum diripuit

1537 Multorum autem hominum vidit urbes et mentem cognovit 1551 Multorum autem hominum vidit urbes et muros cognovit 1561 Multorum autem hominum vidit urbes et mentem cognovit 1567 Multorum autem hominum urbes vidit et mores novit

1537 Multos autem hie in mari passus est angustias proprio in animo 1551 Plurimos vero ille in Ponto passus est dolores suo in animo

1561 Plurimos vero ille in mari passus est dolores suo in animo

1567 Plurimos vero ille in mari passus est dolores suo in animo

1537 Liberans propriamque animam et reditum sociorum 1551 Magna cura servans suamque animam et reditum sociorum 1561 Magna cura servans suamque animam et reditum sociorum 1567 Magna cura servans suamque animam et reditum sociorum

1537 Sed neque sic socios liberavit cupiens quamvis 1551 Sed neque sic servavit soeios tametsi cupidus

1 56 Sed neque sic servavit soeios tametsi cupidus

1 567 Sed neque sic eripuit socios tametsi cupidus

1537 Ipsorum enim propriis stultitiis perierunt

1 55 Sua autem ipsorum insipientia perierunt 1561 Sua enim ipsorum insipientia perierunt 1567 Suis enim ipsorum nequitiis perierunt

1537 Fatui qui boves Hyperonidae Solis 1551 Stulti qui boves super gradientis solis 1561 Stulti qui boves sublimis solis Robin Sowerby 179

1567 Stulti qui boves supergradientis solis

1537 Comederunt, sed hie his abstulit reditus diem 1551 Comederunt, ast is his abstulit reditus diem 1561 Comederunt, ille vero eis abstulit reditus diem 1567 Comederunt, ast is his abstulit reditus diem

This is too small a sample to draw clear conclusions but the comparison suggests that, by 1567, much of the peculiar vocabulary and many of the awkwardly unidiomatic expressions of Divus had been eliminated. However, most of the work had already been done before 1567 by Sebastianus in 1551, whose Odyssey to judge from this small sample is better than Vuolphius' Iliad, and by Castalio in 1561. The one clear improvement in 1567, mores novit, which, unlike mentem cognovit, is idiomatic and makes agreeable sense, is likely to have been anticipated in the 1551 version, where muros is probably a misreading on the part of Oporinus' compositor for mores, and seen as such by the reviser of 1567. The preface of Crispinus to the Odyssey in 1567 signals a new concern for proprietas Latina in the version. This concern is manifested in the second and third sedecimo Genevan editions of the Iliad in 1570 and 1580. In the preface to the 1570 Iliad, the secunda editio,"^^ Crispinus is on the defensive: Ceterum quam apte in hoc eYxeipi6ia) excudendo, et Latina ad verbum versione ex doctissimorum professorum interpretatione concinnanda feliciter versati simus, eorum esto indicium qui sine invidia aut malevolo supercilio de rebus iudicant. In the 1580 edition, the postrema editio ,'^ published after his death, his son-in-law Eustace Vignon, who had inherited his printing business, produces a formulation which defines very precisely the task of the skilful interpreter: est enim sciti interpretis non verborum numerum & ordinem sectari, sed res ipsas & sententias attente perpendere , easque verbis & formulis orationis vestire idoneis , & aptis ei linguae in quem convertitur ("it is the part of every knowing and judiciall interpreter not to follow the number and order of words but the materiall things themselves, and sentences to weigh diligently, and to clothe and adome them with words and such a stile and form of Oration as are most apt for the language into which they are converted").'*^ Then looking back upon the history of versions of this kind, Vignon defends the crudity of the earlier ones (their ruditas) on the grounds that those who made them had the humane desire to accommodate themselves to the needs of tyrones whom they wished to help. Then others, realising that their efforts in this kind were not unwelcome to students, worked all the harder to make the fruits of

^^ 'Hptoim. 'OnTipOD 'I^id<;, Homeri Ilias, id est, de rebus ad Troiam gestis, Latine ad verbum exposita. Secunda editio (apud Crispinum: [Geneva] 1570). ^ 'HpcoiKot. 'O^Tipou 'IX,id(;, Homeri Ilias, postrema editio . . . Latine omnia ad verbum exposita, eta F. Porto Cretensi innumeris in locis emendata ([Geneva] 1580). ^^ The translation is by George Chapman, in the 1611 pr

in Chapman' s Homer, ed. by A. Nicoll (London 1957) I 17. 1 80 Illinois Classical Studies 21(1 996) their labours more apparent. And so many versions of this kind not only from other hands but also from his printing press saw the light, among which was the version of Homer, but one that was not well corrected {sed parum emendata). But at last he has been offered the version which Franciscus Portus revised and cleaned of many errors, ea quam Fr. Partus Cretensis recognovit, & a plurimis mendis repurgavit. Franciscus Portus (1511-1581), a Cretan by birth, had taught Greek in Italy before settling in Geneva in 1562 where he was professor of Greek. In this 1580 preface to the Iliad with Portus' version, Vignon effectively dismisses the second edition published by his father-in-law in 1570, though its reviser may be said to have initiated the drive to proprietas Latina.

Where previous versions of 1. 8 (xiq x' ap a(p(oe Oetov epi5i fyivirymz lidxeaGai;) reproduce the syntax of the Greek as in Divus' commisit pugnare, in 1570 we have Quisnam ipsos Deorum liti commisit ut pugnarentl But the drive to proprietas Latina is only half-hearted. At line

18 •ufiiv |iev Geol 8oTev . . . eKTtepaai npid|ioio 7u6X,iv remains unrevised in the Latin prose: Vobis quidem dii dent . . . expugnare Priami civitatem. And with lines like Et navibus dux fuerat Achivorum Ilium ad (71), translating Kal vrieoa' fiyriaax' 'Axaicov "lA-iov eiaco, there is no seriously sustained attempt to escape the bonds of a rigorously ad verbum principle in the interests of a good Latin style. In other respects, there is no consistent pattern in the revision of 1570. Good and bad are altered quite arbitrarily. In line 3 inferis misit of Divus, translating "Ai5i 7ipoia\|/ev, is improved to Plutoni praemisit, but in line 6 divisi sunt contendentes, translating 6iaaTf|triv epiaavxe, is weakened to dissenserunt litigantes . There are infelicities of various kinds throughout. When Chryses comes liberaturusque filiam ferensque infinita precia liberationis (13), there is the redundancy of laboured verbal explication. When Agamemnon dismisses Chryses, gravem et sermonem mandavit (25) is not good classical idiom. The choice of domicilia (18) for 8(o^iaxa (of the gods) is inappropriate. When Apollo comes down ex caeli verticibus (44), where caeli renders Ou^-u^itioio, the sense is absurd, and when he shoots arrows at the Greeks (Pd^A-e), the choice of the verb iecit (52) is again absurd and its sense unhelpful. These infelicities are ironed out in the Latin of Franciscus Portus in his revision of 1580: redempturusque filiam, ferens immensum pretium (13), dura autem mandata dedit (25), caelestos domos (18), Olympi de vertice (44) and feriebat (52). It is clear that the interpreter here has a much better grip on both the Greek and the Latin. In countless instances, the version is improved by the use of an idiomatic phrase; as, for example, when Hera puts it in Achilles' mind to call the council, in animo posuit (55) improves upon the version of 1570, with in mentibus posuit for eiti (ppeai OfiKe; iratus animo (44) is idiomatically easier than iratus cor (1510) for xcoojievoq Ktip. In these instances, Portus abandons religious adherence to an ad verbum principle in the interests of an idiomatic version that will read well in Latin. 1

Robin Sowerby 1 8

Like his predecessors, in making his revision Portus has consulted the Greek interpretatio in the scholia. In the opening Hnes, his Latin seems to derive from Greek glosses in the following instances: consilium (5) from

yvcbiiri for ^o\iki\ ; disiuncti sunt (6) from SiaxwpioGriaav for 5iaoTr|Tnv; nobilis (7) from eijyevTic; for Sioq; immisit (10) from evePaA,ev for «pae; pestilentem (10) from A,oi|iiktiv for KaKT|v; redempturus (13) from A.-UTpwao^evoq for ^vaofxevoq. But, as might be expected, an established scholar like Portus is much more discriminating in his use of glosses than a novice like Vuolphius. Only in one instance has he definitely replaced a good literal rendering {suscitavit for cbpae) with an inferior gloss and even here the Latin makes good sense. In the opening lines of Portus' version, a slight departure from the literal Greek results from sensitivity to the Latin

derived from the Greek and suggested by it. Despite the scholiast's note drawing attention to the force of the prepositional prefix in 7i;poia\|/ev (3), he retained the earlier reading of Vuolphius, orco demisit, perhaps recalling a Virgilian echo of this line of Homer at Aeneid 9. 527: quern quisque virum demiserit Orco. In the next line, the change certainly results from this process of recall:

a{)xo\)(; 5e eXcopia xzvxz icuveaaiv oiMvovai xe naai ^

1570 ipsos autem laniamenta fecit canibus Avibus omnibus

1580 ipsos autem praedam dilaniandam fecit canibus Alitibusque omnibus

Virgil canibus data praeda Latinis Alitibusque jaces.^^

We may believe that for Portus the task was not entirely mechanical; here is an intimation of a genuine poetic sensibility. The revision by Portus in 1580 was a great improvement on the revision of 1570 and, as a result, his is a better version than any that had been made previously. Yet there are various insufficiences. At 14 longejaculantis of Divus is retained. At line 24 aXk' otjk 'Atpe"i5r| 'Ayaiieiivovi livSave 9\)^w is garbled in the Latin rendering: At non Atridae Agamemnonis placuit omnino. This last word may be a typographical error for animo, but on average there is a serious error of some kind, typographical or other, every fifteen lines or so. At 35-36,

nokXa. 5' eTceu' dmve-uGe kvcov fipaG' 6 yepavoq 'A7t6A,X,a)vi avaKTi

multum deinde procul inter eundem precabatur senex Apollinem regem,

*^ Virgil, Aen . 9. 485-86. 182 Illinois Classical Studies 21(1 996)

it is more difficult to divine the original of which this might be a corruption.

At line 71, the 1570 version above is at least fairly recognisable as Latin. It is difficult to say the same of the rendering of this line in Portus: et navibus ductus fuit Achivorum Ilium intra. As this example suggests, the drive to proprietas Latina is not systematically sustained in Portus either. When Chryses asks Achilles to swear an oath that he will help him in words and deeds (76-77), while the reviser of 1570 had recast the Greek in order to write good Latin, Portus produces an ad verbum rendering that has no proper Latin construction:

Ka{ |j.oi 6|ioaaov

f| |iev ^01 npocppcov eneaiv Kal xep

1570 etmihiiura omnino te mihi verbis et manibus opem laturum

1580 etmihiiura certe mihi quidem promptus verbis et manibus auxiliari

These inconsistencies of practice may be the result of insufficient application or of an indecisiveness about the function which the version might best serve. The version of Portus is no longer a clavis, nor, though it shows what might be done, is it consistently sound Latin; it is a hybrid form, neither one thing nor the other. In his magnificent folio edition of 1583, the young French humanist

Jean de Sponde promises Homer cum latina versione . . . emendatissima aliquot locis iam castigatiore.'^'^ This is an important edition as it is the only complete edition of the poems in the whole period from the first edition to the eighteenth century to be accompanied by a commentary, a distillation in Latin of the Greek tradition, which was used by translators from Chapman to Pope. Spondanus used the 1570 version as his base (perhaps because he started work earlier before the 1580 version was available). About one third of the lines are unchanged. Sometimes he retains inferior renderings from 1570 such as the long-winded and tautologous liberaturusque filiam et ferens infinita precia liberationis (20), longejaculantis (14), caelestia domicilia (18), durumque sermonem mandavit (25), ex caeli verticibus (44). Occasionally he adds something of his own as in ante tempus to render the

prefix in Ttpoiayev (3), capessantem for avxiocoaav (33), not a felicitous rendering, iniuriose for KaKox; dcpiei (25) and in praecordiis for cppeai, of Achilles (55), which unlike animo retains the plural and unlike mentibus is idiomatic. He had before him the version of Portus, which in some lines he

substitutes for 1570 and in others he amalgamates with it. It is possible to see definite patterns in his use of the version of Portus. He takes the

''^ Homeri quae extant omnia . . . cum latina versione omnium quae circumferuntur

emendatiss. aliquot locis iam castigatiore . . . perpetuis . . . in Iliada simul et Odysseam J.

Spondani . . . commentariis (Eusebii Episcopii opera et impensa: Basileae 1583). Robin Sowerby 183 occasional phrase for clarity's sake, as when Achilles suggests to

Agamemnon that a priest be consulted to see if Apollo is angry because of

£.<}X(o'kr\q (65), rendered in 1570 by ob vota and by Portus as ob vota non reddita. He preserves a more conservative rendering which had been changed in 1570, for example reverting to nidorem (66), "smoke from a sacrifice," for Kviar|(;, which in 1570 had been incorrectly rendered by the rare word arvinam meaning "fat." Thirdly and most frequently he chooses the expression of 1580 when it is more forceful, as in disiuncti sunt (6), contumelia ajfecerat (1 \),feriebat (52) and interrogemus (62). He does not reject the movement towards proprietas Latina, retaining commisit ut pugnarent (8), and improving upon both 1570 and Portus in the following lines previously quoted: Et navibus dux fuit Achivorum ad Ilium (71) and et mihi iura I certe quod mihi promptum verbis et manibus te auxiliaturum (16-17). On the other hand, he rejects the freer recasting of the two Genevan versions as in the case of 76-77 quoted above, where he is more literal than the 1570 version, and in cases where Portus has changed tenses in the interests of fluency or variety. In line 12, when Chryses came to the ships of the Achaeans to ransom his daughter, Portus translates T|^0e as venerat, and when he prays that the Greeks will free her reverencing Apollo, he translates a^o^evoi (21) as veriti. Spondanus never follows either version in such deviations, preferring instead a literal rendering wherever possible. He endeavours to write sense and to be as literal as possible, eschewing any tendency to elegance. He is more systematic than Portus in his own revision and more consistent in his practice so that he produces a version that is less of a hybrid. In some ways this is the high point of the versio, particularly since it is typographically sound and generally well punctuated, improving in these respects on most of his predecessors. Spondanus is one of the very few editors of Homer in the Renaissance and the seventeenth century to give the Greek text, in his attention to the versio and in his annotations, the steady and conscientious application that it both needed and deserved. In 1589, with the publication of the sedecimo edition of Henricus Stephanus,"*^ the history of the Latin version took a curious twist. On the title page Stephanus promises Homeri poemata duo . . . cum interpretatione

Lat. ad verbum , post alias omnes editiones repurgata plurimis erroribus, (et

quidem crassis alicubi) partim ab Henrico Stephano , partim ab aliis ut te epistola ad lectorem docebit. But the epistle to the reader tells a very confusing story. In his preface, Stephanus feels the need to account for the delay in the appearance of the edition, saying that he has been away and working on other projects. He asks whether in the meantime the appearance of any

^ Homeri poemata duo . . . cum interpretatione Lat. ad verbum , post alias omnes editiones repurgata plurimis erroribus, ( & quidem crassis alicubi) partim ab Henr. Stephano , partim ab aliis etc. ([Geneva] 1589). 184 Illinois Classical Studies 21 (1996)

other edition might have saved him the trouble. It is slightly surprising that he then mentions the edition of Giphanius published in Strasbourg seventeen years earlier in 1572.'*^ This edition he dismisses out of hand, noting Giphanius' admission that for the Greek text he had simply reprinted the text of Stephanus' own landmark edition of 1566. Perhaps Stephanus refers to this edition partly for reasons of self-advertisement. As for the Latin version used by Giphanius, Stephanus points to what he calls an example of such crass ignorance on its very threshold that the reader can have scarcely any expectations of it. For the words lAIAAOI OMHPOY H A

PAH^QAlA are rendered in Latin Iliados Homeri aut i compositio , where the

Greek article (r\ ) has been mistaken for the particle fi , "or," and the genitive form Iliados is also wrong. The whole version, as Giphanius states quite openly, is taken from the edition of Crispinus in 1570, together with this unfortunate error (corrected by Portus in 1580). Since the 1570 version contains some intelligent revisions, we must conclude that the headnote was probably a crudely botched job on the part of an ignorant compositor which was not spotted by either the reviser, who may have had nothing to do with the actual process of printing, or by Crispinus. There is no mention in Stephanus' preface of the more recent edition of Spondanus, prompting the thought that this preface may have been written considerably earlier than its date of publication in 1589. When Stephanus finally comes to fulfil the promise of his title page, what he says about the composition of the version is very difficult to comprehend (sig. 80:

Hoc enim tantum dico. Si quis quam multa etiam post Franciscum Portum, et quidem in ipso etiam Fr. Porto emendata fuerint, atque adeo emendari debuerint, perpendat, fassurum esse, non iniuria me interpretatione illius acquiescere noluisse: etiamsi in primo Iliadis libro eam magna ex parte sequutus essem. Quum enim officina typographica ilium primum librum ad finem perduxisset, ego attentius interpretadonem

illius examinans, eam illi remisi, et ut recognosceret rogavi. Verum quod ab illo postulabam, et obtinere non poteram, tandem partim a me ipso, partim ab ipso postulare et impetrare necesse habui.

I have only this to say. If anyone weighs up how many errors even after Franciscus Portus and indeed in Franciscus Portus himself might be

corrected and indeed ought to be corrected, he will acknowledge that it is

not without justice that I was unable to acquiesce in his interpretation even

if in the first book of the Iliad I followed it for the most part. But when the printing office had gone through to the end of the first book, looking over

his interpretation more closely, I first sent it back to him and asked that he

revise it. But because I could not get what I was requesting from him, in

the end I decided it necessary to ask and demand it partly from myself and partly from himself.

'*^ 'OnTipou 'IA,iaq t\ \iakXov (XTiavta ta a(p^6|ieva, Homeri llias, seu potius omnia eius

quae extant opera . Studio et cura Ob. Giphanii etc. (Theodosius Rihelius: Argentorati 1572). Robin Sowerby 185

It appears from this that Stephanus had originally used the version of Portus, published in 1580, as his base for revision, if indeed he had revised at all.

Only when Book 1 had come back from the printer had he rightly realised that it needed systematic revision. So he sent it back to Portus for more work. But herein lies a problem, for Portus had in fact died in 1581, only a year after the publication of his version, and since he was resident professor of Greek in Geneva where Stephanus also had his printing press and from where the 1589 edition seems to have been published (there is no indication of place of publication in the book; some catalogues give Paris) it is difficult to suppose that Stephanus, even if at this late stage of his career he had embarked upon his nomadic existence, could have been unaware of his death. In the Latin quoted, it may be that in obtinere non poteram Portus' death is to be inferred, but this is a very strained reading. In the Genevan edition of 1570, Crispinus had published a brief introduction to the Homeric poems written by Portus with the prefatory exhortation: Verum ipsum Portum praefantem, quum secundum Iliadis interpretationem publice aggressus est, audiamus . It seems likely from this that Portus had a versio as early as 1570 for the purposes of public lectures, i.e. for explication of the text in the tradition of Leontius and Callistus, and that Stephanus had seen it in manuscript. This might account for what otherwise is a curious distinction between post Franciscum Portum, which might refer to the general enlightenment afforded by his public explication of the text, and in Fr. Porto, which must refer to his actual version itself. In this case Stephanus' own preface, published in 1589, may well have been written years earlier. This might account for the lengthy discussion of the edition of Giphanius of 1572 and the failure to mention Spondanus of 1583. But there remains the problem of the final sentence. What can ab ipso mean?

Perhaps it might be understood to mean from Portus' version, that is,

Stephanus felt it necessary to seek changes both from his own hand and from the version of Portus that he had used originally. In saying initially that he had followed Portus for the most part, he does allow the implication

that he had another versio. Perhaps the meaning is that after Book 1 returned from the printer, he consulted Portus more thoroughly a second

time. This interpretation is very strained indeed, but it does have the merit

that it fits the facts of the case, for it is most surprising to find that in Book 1 Stephanus has used as his base the much inferior version of Vuolphius of 1551, to which he did indeed make changes himself with help from the 1580 version of Portus. More than any previous reviser, Stephanus boldly departs from the form of the Greek: Ai6(; 5' kxzKzxzxo |3o-uA,r| (5) is rendered by ac lovis consilium exitum habuit, and in the previous line he keeps the more elegant Latin of Portus, praedam dilaniandam. Some of his alterations (tending as here to abstraction of the concrete and physical) move away from the simplicity of Homer's style in the interests of good Latin, introducing elegant touches or emphatic expression beyond anything in Spondanus. He 186 Illinois Classical Studies 21 (1996) writes, for example, preces fundebat (35) for precabatur (1551, 1580), habet imperium (79) for imperat (1551) or dominatur {\5%Q), ut adfinem perducat (82) for donee perfecerit (1551, 1580), eum summa fiducia (85) for confisus valde (155 1, 1580). He also makes a number of changes which are not so much for elegance as to improve the clarity of the Latin so that it makes sense more easily, changing, for example, his vero surrexit (68) to inter illos vero surrexit. The addition of illius gives a clear construction to the following: sonuerunt vero tela in humeris illius irati (46). Similar is the addition of est in the line potentior enim rex est quum irascitur viro inferiori (80). The addition of viae clarifies the role of Calchas: et navibus viae dux fuerat Graecorum ad Ilium (71). The author of the Thesaurus Graecae Linguae (1572), whose father Robert had produced the Thesaurus Linguae

Latinae (1532), was ideally equipped for the task, if he had chosen to give it the attention it needed. Unfortunately he made the task more difficult for himself by his inexplicable decision to use the version of 1551 as his base. He takes idiomatic expressions from Portus, as in dura mandata dedit (25), but there are many traces of the unidiomatic and cumbersome expressions of 1551. Why did Stephanus leave commisit pugnare (8) unamended, when he obviously did not conceive of the versio in the narrowest terms as an ad verbum clavisl He retains depopulari Priami urbem (19) instead of the better expugnare (15S0), jaculis (42) for arrows, and priusquam ipsam senium invadit (29) and nocti adsimilatus for the simpler and more natural expressions antequam ipsam senectus adeat and nocti similis. The combination of a move to elegance with the ungainly expression of 1551 sometimes produces a portentous effect. In 1580, where Portus retained the simple version of Divus at line 85, confisus valde die vaticinium quod scis, Stephanus has tu summa cum fiducia dicito quodcumque novisti. Here the first half is his own change and the second comes from 1551. He occasionally leaves odd additional phrases from 1551 like undique in qui Chrysen undique tueris (37) and per vices in ilium vero per vices respondens allocutus est pedibus celer Achilles (84). The problem with Stephanus' version, therefore, is not so much what he changes as what he leaves unchanged. Nor does he seem to have made much use of the scholia, or he would have had better renderings for orco demisit (3) and dissenserunt contendentes (6). A change early in the first book that does result from a glance at the scholia only serves to suggest that Stephanus' mind was not fully engaged in the task of revision. When Agamemnon says that he will take Chryseis home to Argos, the scholiast explains that Argos is in the Peloponnese, a gloss perhaps helpful to a Byzantine pupil or to a beginner in 1589. But to substitute in Peloponneso (30) for Argos in the version

itself (it is in no other before and not generally repeated) can only cause confusion. Although Stephanus has naturally eliminated the gross errors of 1551, some oddities remain and the resulting confection is as much of a hybrid as previous efforts. In fact Stephanus has introduced inconsistencies of more kinds and on a greater scale than before, yet such was his authority Robin Sowerby 187 that the 1589 version was reprinted more often than any other during the next hundred years, as the stemma (below, page 189) shows. Indeed it was Stephanus' version rather than his father's that was revised by the younger Portus for the Genevan edition in sedecimo of 1609.^^ Aemilius Portus had his father's version of 1580, and his general tendency is to follow his father's use of the scholia. On the title page are the words paternos commentarios secutus , so that Aemilius had access to material probably used by his father in his public lectures on the text. His revisions are made to a more consistent pattern than those of Stephanus. He reverses the trend of Stephanus away from the Greek, changing exitum habuit (5) back to perficiebatur . He was not interested in elegance, nor was he primarily concerned with proprietas Latina. His main concern was the traditional one, to give a straightforward explication of Homer's words. The wordy gloss on Kpoia\|/ev (3), orco ante-iustum-tempus-cum-detrimento demisit, shows his approach at its most pedantic. Nevertheless, some of his choices (or failures to make a change) show the weakness of his predecessors. He preferred dissenserunt (6) to disiuncti sunt in 1580 and he retained the weak gloss of Stephanus in Peloponneso (30). Once again the revision is rather hit and miss. He followed the practice of Stephanus in putting brackets around words that he adds to make the sense clearer. When the Greeks acclaim Chryses saying that he should be respected, aiSeioOai 0' lepfia (23) is rendered not by a gerund as in other versions but by revereri [oportere]; thus the version is a key, but good Latin sense is preserved. Yet as the rendering commisit pugnare (8) shows, he was not fully consistent in this practice. Nevertheless, while the version of Stephanus shows what might have been possible with application and some feeling for style, the version of Aemilius Portus is second only to Spondanus in offering an old-fashioned unpretentious key to the Greek. Neither of these versions, however, was used as the base for the next revision in what was another landmark edition of Homer emanating in 1656 from the press of the distinguished Dutch printer Johan Hack: Homed Ilias et Odyssea , et in easdem scholia , sive interpretatio Didymi. Cum Latina versione accuratissima . . . Accurante Corn. Schrevelio.^^ This is the first edition of Homer in which the scholia are printed with the text in a form that allows them to be easily read. In conception and appearance, this edition set a new standard. The Greek text is produced in a fine typeface with the scholia below and the Latin version set to the right on the same page in reduced type. With due decorum, the Greek is given a new prominence.

^° Heroica. Homeri Ilias, cum Aemilii Porti, Francisci Porti Cretensis F . Latina ad verbum interpretatione ; quam is, paternos commentarios accurate sequutus, ab innumeris mendis repurgavit etc. (per lohan Vignon: Aureliae Allobrogum 1609). ^' 'OnTipou Wmc, KoX '05iL)oaeia, kui tic, a\)xy\\ oxoXia, r\ ii,r\yr\a\(:, AiSij^od, Homeri Ilias et Odyssea, et in easdem scholia, sive interpretatione Didymi. Cum Latina versione accuratissima etc. (apud Franciscum Hackium 1656). There is a duplicate of this published at Amsterdam in the same year with the imprint "ex officina Elzeviriana." 1 88 Illinois Classical Studies 21(1 996)

while the reader is presented with the two most useful aids to interpretation immediately adjacent and in a form that is aesthetically pleasing to the eye. Schrevelius in his preface professes confidence that the learned world will prefer this edition before all others nitore et diligentia castigationis , for its handsomeness and for the diligence with which it has been prepared. He says that the Greek text is from the best editions of Tumebus and Stephanus while the Latin version is mainly that of Giphanius, though in many places changed and corrected with reference to the versions of Stephanus and others. Since Giphanius reprints without revising the version of 1570 published by Crispinus, this is a disappointing choice and one that would not have been made by a scholar who had seriously investigated alternatives. Despite Schrevelius' claim about revisions, in the first book of the Iliad he has made very few changes, none of which comes from the version of Stephanus. His favourite alternative to 1570 is the unrevised version of Divus (printed variously in Geneva and Basle). He has also taken one or two renderings from the generally better version of Portus in

1580. It is difficult to discern any real principle at work in his tinkerings but he certainly has no concern for Latin elegance or even propriety, rendering the line ev0' akXoi [ikv Kavteq e7i:e\)(pT||j,Tiaav 'Axaio{ (376) as tunc alii quidem omnes in idfaverent ore Achivi. This is one of the few places in

Book 1, about twenty in all, where he has produced a new rendering of his own. Strangely the same line at 22 is unrevised from 1570: tunc alii quidem omnes faverunt Unguis Achivi. Having chosen an inferior version to start from, he then introduces new horrors himself. Some of these may be typographical. When the story is told how Zeus threw Hephaestus from Olympus, dejecit pede prehensum e caelo (vero lumine) divino (591), lumine glossing Prj^ot), "threshold," is probably a misprint for limine. But when he attempts to explain the difficult word nz\x.n^^oXa (the five- pronged forks on which the flesh or the innards of the beast were roasted after the sacrifice) the failing is more radical: iuvenesque secus ipsum tendebant verva quinque-ordinum (463) rendering veoi 8e Tcap' av)t6v exov 7te)X7i(opoX,a xepaiv is nonsensical. This edition was immediately subjected to detailed criticism by Meric Casaubon, son of Isaac and a notable scholar in his own right, in a lengthy Dissertatio entitled De nupera Homeri editione Lugduno-Batavica, Hackiana, cum Latina versione etc. (1659).^^ Casaubon seems to have been working on the Odyssey when the Dutch edition was published. Most of his specific discussion draws upon examples from the Odyssey and in his review of previous editions he has probably consulted editions of the Odyssey in the first instance. But he has examined the Dutch edition in toto and his general remarks and conclusions can apply equally to both poems.^^

Having noticed errors (his word is portenta) in a rapid review of the versio,

^^ See above, note 2. ^^ What follows summarises pages 9-10 of Casaubon's Dissertatio. .

Robin Sowerby 189

[Leontius Pilatus — __^ ~~ Florence, c. 1360] Andreas DIVUS (1537) 1537 Venetiis (Latin only) I 1538 Parisiis (Latin only)

I 1538 Lugduni (Latin only) Joannes VUOLPHIUS (1551) 1560 apud Crispinum [Geneva] 1551 Basileae 1561 Castalio, Basileae 1567 Castalio, Basileae

ANON. (1570) 1570 apud Crispinum [Geneva] 1572 Giphanius, Argentorati Franciscus PORTUS (1580) 1582 Castalio, Basileae 1580 [Geneva] 1606 Lectius, Aureliae AUobrogum 1621 Genevae 1641 Genevae

Johannes SPONDANUS (1583) 1583 Basileae 1606 Aureliae AUobrogum Henricus STEPHANUS (1589) 1686 Basileae 1589 [Geneva] 1604 Paulus Stephanus [Geneva] 1617 Genevae Aemilius PORTUS (1609) 1621 Paulus Stephanus [Geneva] 1609 Aureliae AUobrogum 1622 I.T.P., Parisiis 1629 Genevae

1648 I.T.P., Amstelodami 1648 Cantabrigiae 1650 Amstelodami 1664Cantabrigiae Cornelius SCHREVELIUS (1656) 1672Cantabrigiae 1656 Lugduni Batavorum 1679 Cantabrigiae 1656 Amstelodami 1686 Cantabrigiae

ANON. (1689) 1689 Cantabrigiae

The Versio Latina of Homer's Iliad

A stemma showing the relationship and incidence of the Versio Latina in printed editions of the Iliad in the sixteenth and seventeenth centuries, derived from a collation of Book 1 1 90 Illinois Classical Studies 21(1 996) he estimates that there are at least six hundred in all. He is talking here about glaring mistakes of interpretation rather than inelegancies of style. He then tells us that he undertook a review of previous editions. He was amazed to find that the errors of Schrevelius do not appear in an earlier version, which, he says, had been quite widely disseminated and previously known to him (it had been printed in Cambridge in 1648), that of Aemilius and Franciscus Portus. (Aemilius Portus had included his father's name on the title page.) The implication here is that with a little effort Schrevelius could have found a better version from which to start. He then consulted the edition of Castalio of 1567, which he found to be not as pure as Portus but better by far than Schrevelius. (For the Odyssey, Castalio had revised the 1551 edition; for the Iliad, he had reprinted Divus unrevised.) The worst versions of all he had seen were those of Giphanius (reprints of 1570, used by Schrevelius as his base). Commenting on the Amsterdam editions (of 1648 and 1650, both containing reprints of Stephanus) he notes that most of the ghastly errors of Schrevelius do not appear in these versions but that they contain many inexcusable mistakes that are correct in Schrevelius: Amstelodamensis autem, quamvis multa correcta exhibeat, quae peccat enormiter Hackiana: at ilia vicissim plura, aut certe non pauciora admittit, quae piatonis non minus egeant, quae in Hackiana recte habeant.^'^ He notes, therefore, that error in the versio was quite random and that there had not been a progressive elimination and improvement. After making his review, he asks how this edition of 1656, which prides itself on its elegance, could ever have been published tot versionis ulceribus foeda ac deformis.

To cap it all, he finds that the scholia, reprinted from the Basle edition of 1539 and the chief source of the publisher's pride, are full of typographical errors. Like the first Graeco-Latin edition published in Basle in 1551, this edition which appeared to mark an advance in Homeric scholarship actually reveals the opposite, a regressive ignorance of Homer in the culture of the times. His review of previous editions of Homer led Casaubon to the conclusion that the edition of Schrevelius was not only a particularly bad case in itself but was also symptomatic of a deeper malaise in Homeric studies more generally in the Renaissance and the seventeenth century (8):

Cum enim aliorum in quocunque genere scriptorum Graecorum versiones extent haud paucae, quas merito laudamus: etiam illorum, in quibus labor longe major et difficultas ingenia non vulgaria deterrere poterat: quis non miretur fati Homerici iniquitatem, cui cum palmam dent omnes ingenii, et omni laude, longissimo intervallo praecellentem agnoscant: nondum tamen repertum esse, qui tot insulsissimi interpretis hallucinationibus et barbarismis horrentem et squallidum, et quidvis potius appellandum quam Homerum; meliore cultu omatum et politum, Latinis auribus proponeret?

^^ Casaubon (above, note 2) 10. For the Amsterdam editions, see the stemma (previous page). Robin Sowerby 191

Since there are many versions of other Greek writers of various kinds which we rightly praise, and even ones in which the much greater labour and difficulty could put off scholars of rare intellect, who will not marvel at the unfairness of Homer's fate, for when all people of intellect give him

the palm and recognise that he is pre-eminent above others by the greatest degree and when he is obscured and made ghastly in appearance by so many botchings and barbarisms in a tasteless interpretation and ought to be called anything but Homer, there is no one to be found who can bring him to our Latin ears adorned and refined with greater polish?

He was concerned for the effect that the versio might have on those tyrones it was designed to help (10):

Nee adeo tamen Homed ipsius injuria commovebar, quem nemo, qui sapit, ex versionibus vel accuratissimis, nedum vulgaribus, aestimabit: sed tyronum praecipue intuitu, quorum teneros animos, cum amore et admiratione Poetae incomparabilis imbui, reipublicae literariae et posteritatis adeo intersit: mirum ni tot impactae cruces, et tenebrae offusae, ubi Graeca plerunque liquidissima sunt et elegantissima, vel de proprio cogant ingenio desperare; vel alieno (Homerum intelligimus) immerito obtrectare.

Nor am I so much moved by the injury to Homer, whom no one who has sense will judge from versions even the most accurate, let alone the ones in common circulation, as by consideration for beginners, for it is important to literary culture and posterity that their young minds be inspired with a love and admiration for this incomparable poet, yet it would be a miracle if so many knotty cruces and such general obscurity, where the Greek is most clear and elegant, should not compel them either to despair of their own intelligence or unjustly to find fault with Homer's.

Appalled by the errors and obfuscations of the Latin version, Casaubon had a clear idea of what was necessary for an improvement and provides examples of his own in the hope of stimulating some other scholar to undertake the task. Although he is unequivocal in his praise for Aemilius

Portus, he still felt that there was room for further improvement in the understanding of Homer and his recommendation of the best version comes with a warning to any reader who is about to use it: In any places he happens to be in difficulty or doubtful, he should first consult the scholia or the commentary of Eustathius before pronouncing upon it.^^ Among the examples that follow is discussion of a phrase from Odyssey 3. 340 which also occurs at Iliad 1 . 47 1 : eTiap^djievoi Sendeaaiv . Casaubon notes that Hesychius gives the meaning OTteiaavteq, libantes, and Didymus has the long gloss, e7iap^d|ievoi- xov Ttiveiv dpxriv TtoiTiaavxeq, Kal d7tap^d|ievoi Kttl cneiaavxec,, while Portus has iterum exorsi (and others have incipientes in their versions). He prefers the translation libantes poculis or at least

' Casaubon (above, note 2) 52. 1 92 Illinois Classical Studies 21(1 996) auspicati a poculis, where auspicor, "begin," carries the connotation of good omen, therefore fitting the religious context, for the Greeks are propitiating Apollo. Modern commentators still differ in their exact interpretation of the Homeric phrase, but Casaubon's version does have the merit of making good sense, whereas incipientes, the rendering of Divus, repeated in 1570 and therefore in Schrevelius, begs the question about what happens after this beginning (470-72):

Juvenes quidem crateras coronaverunt vino Distribueruntque omnibus incipientes poculis Hi autem totam diem cantu deum placebant

Portus in 1580 wrote iterum exorsi, where the latter word renders the tense of the participle but the former introduces a further confusion. Casaubon's Dissertatio proves not only that a much better versio was a desired contemporary need but that it was well within the capacity of the scholars of the time. His challenge was not seriously taken up, but the versio was revised once more in the seventeenth century for the Cambridge edition of 1689^^ by a scholar who had read his Dissertatio, to which he refers in his preface. The reading of Casaubon made him the first to render eTtap^d^evoi Sercdeaaiv (471) with the interpretation of the scholia in mind: Distribuerunt omnibus qui poculis libabunt auspicantes. However, in Homer it is the young men who do the libating (perhaps only by gesturing with the cup as they hand each one round). The three portenta of Schrevelius discussed above are corrected. The rendering of TteiiTtcoPoAxx is not elegant but shows a desire to write self-explanatory sense: juvenesque secus ilium tenebant veruta in quinque mucrones fissa manibus (463). The base from which the editor is working, despite Casaubon's advice, seems to have been Portus pere rather than Portus fils. He has consulted others, including Spondanus, which had been reprinted in Basle three years earlier, and the second edition of Crispinus in 1570. He may have preferred the elder Portus because of his tendency (reversed by his son) towards something stylish, for the 1689 version, more than any other since Stephanus, departs from the exact form of the Greek in the interests of an elegant Latinity. The ablative absolute orta contentione rendering epiaavte (6) and the changes in tense from r\XQe (12) and Xiaaexo (15) to venerat and oravit introduce the Roman logic of time. The additional supplex, only in 1689, in supplex oravit sounds like good Latin. There are attempts at Roman idiom and emphasis in sexcentos for ocTrepeiaia (13) and noxium for KttKriv (9). Since he has a better base than Stephanus and has been more systematic (Stephanus inexplicably left commisit pugnare from earlier renderings) the Cambridge editor in 1689 has produced a version that is

insuper in hoc volumine . . . Iliadis nova interpretatio Latina etc. (J. Hayes: Cantabrigiae [printed]; E. Brewster: Londini 1689). Robin Sowerby 193 much less of an unsatisfactory hybrid than had been the case in 1589. Yet he was not so systematic that he eliminated all infelicities either in the Latin expression, where we still find et navibus dux fuit Achivorum Ilium ad (71), or in the interpretation of the Greek, as in ex caeli verticibus (44), both examples being repeated from 1570. All subsequent revisions of the version in the eighteenth century descend in direct line from this edition, which marks a natural break, as, on the basis of Book 1, it is possible to say that the portenta, whether typographical or other, have now been largely eliminated. Subsequent editors tinker with the Latin but more or less give up any attempt to make the versio more elegant, despite the reasonable working base provided in 1689.^^ The long preface to this edition, whose author is keen to assert his scholarly credentials, contains an interesting section on the versio in five marked stages that has much to suggest about attitudes to the task not only in 1689 but throughout its history. First, there is an acknowledgement of initial reluctance. Since the conrmiissioning editor wished to keep the form of the earlier Dutch edition, the version was a necessary part of the task at hand but the least congenial, for there is little kudos for a good interpreter but much condemnation for one who proves inadequate. And it is particularly difficult in the case of Homer, coming after so many others have tried their hand in a text of such antiquity that is obscure both in content and style and where there is such disagreement about its interpretation among both ancient and modem scholars: in tarn obscura et rerum et verborum antiquitate, atque ilia multiplici et paene infinita grammaticorum veterum et recentium inter se dissensione . But when we were persuaded (the royal "we," presumably) that much could be gained simply by correcting a few places here and there, paucos hie illic locos corrigendo , we undertook this task as well. Previous revisers had been reluctant for the same reasons and had adopted the piecemeal approach as recommended to the Cambridge editor by his commissioning editor or the equivalent of an editorial board.

But at the second stage it was found that the piecemeal approach could not be adequate because the Dutch edition was so bad, both in the editor's experience and as demonstrated in Casaubon's Dissertatio. Something more considerable was needed. The third stage concerns the assembling of aids: the scholia, the commentary of Eustathius, the lexicon of Hesychius,

^' The main eighteenth-century editions are: '0|iripo-u 'lA-iac; Kai '06v)Ooeia, Homeri Ilias etc., vols, et Odyssea . . . cum Latina versione . . . opera, studio et impensis Josuae Barnes 2

(apud Cornelium Crownfield: Cantabrigiae 1711); Homeri Ilias Graece et Latine . . . edidit

Samuel Clarke, 2 vols. (J. J. Knapton: Londini 1727); '0|ifipou "Artavxa, Homeri opera omnia, ex recensione et cum notis S. Clarkii . . . cura J. A. Ernesti etc., 5 vols, (impensis G. Theophili Georgii: Lipsiae 1759-64). Barnes tinkers with the Latin without making it significantly more elegant. Like other editors, Clarke claims that he has been a great corrector, but on the evidence of Book 1 of the Iliad has done little. In his preface he says that he has attempted to make the Latin correspond as closely as possible to the Greek on the ad verbum principle. Ernesti essentially leaves Clarke alone, admitting that the versio is permultis locis parum Latinum aut elegantem. 1 94 Illinois Classical Studies 21(1 996) the thesaurus of Stephanus, the Latin interpretation of Portus (all recommended by Casaubon) and finally, with an inexplicable bathos that threatens to undermine the seriousness of the whole enterprise, Hobbesii nostratis Anglicana, the English version of our very own Hobbes. Hobbes,

"studying poetry as he did mathematics, when it was too late," in Dryden's witty dismissal, ^^ produced his Homeric versions in 1676 at the age of 88. They were reprinted twice and must be evidence of an English interest in and desire for Homer, but what profit a serious scholar could have imagined he might derive from them, it is difficult to say. Hobbes and Eustathius scarcely seem to weigh equally as sources of potential enlightenment. But the editor assures us that he does not have a religious attitude to his authorities; where the situation demands, he will make his own judgement. At the fourth stage comes a thoughtful discussion about the character of the versio, about its function and about the guiding principles of the interpreter:

It is immediately admitted that there are certain ornaments in versions that have no place here. We have not tried to give Latin speech in which you cannot recognise any traces of the Greek. Usefulness rather than elegance has been the goal. But how can any version of Greek poets be useful except to those who can assess their meaning without the help of it? He who does not know Greek, seeks Homer in vain even in the most elegant version: who knows Greek well, does not need one. Accordingly, we have followed the order and structure of the poet's words as far as the Latin

language will allow and our concern has been first to express the primary

sense of any utterance rather than to follow it exactly, since we believe that those who have a litUe learning can most be helped by such a method.

But we have not spumed any elegance that is not incompatible with these guidelines. Those verbal monstrosities that are a result of imitating Greek formations we have banished and we have cleaned away much of the general barbarity in an attempt to prevent tyrones from being alienated once and for all from Homer in such a ghasUy form.

There is an unmistakable allusion to Casaubon here. Writing on the versio in response to the Dissertatio, the Cambridge editor attempts to square the circle and verges on self-contradiction as he endeavours to reconcile incompatible claims. Nevertheless, the general aim is fairly clear and the intention is decent enough. At the fifth and final stage comes a confession of inadequacy. If the three or four opening books seem to be less felicitously done than what follows, the reader must impute the failing to the haste with which he has had to comply with the demands of the press. In the whole enterprise, the end is more correct than the beginning. In sum, initial reluctance, recognition of a task that has so far been inadequately done, the assembling of aids that will lead to improvement, some agonising over the principles to be followed in such an undertaking and finally in the

'^ In the preface to the Fables (1700), in K. James (ed.), The Poems ofJohn Dryden (Oxford 1958) IV 1448. .

Robin Sowerby 195 face of publishers' deadlines confession that more work needed to be done, all this we have encountered before but never so clearly set out or occurring in the one place. This 1689 version was chosen by Jean Boivin, professor of Greek at Paris, as he contemplated his Projet pour une nouvelle edition d' Homere. The edition never came to be, but the proposals survive in manuscript form and have been summarised by Noemi Hepp.^^ Boivin had evidently examined the version in its previous manifestations from Leontius onwards and found the 1689 version to be the best. He proposed to revise it using other versions and according to principles of his own which show a new clarity of thinking about aims and objectives. From the foregoing account, it is evident that the history of the versio is marked by two often contradictory impulses. On the one hand is the tendency most marked in the revisions of Spondanus and Aemilius Portus to concentrate on making good use of the scholia to give an accurate verbal interpretation. On the other hand is the tendency (whether alongside the use of the scholia or not) as in the revisions of 1570, of Franciscus Portus and of Stephanus to make changes in the direction of sound and even elegant Latin. Boivin is the first writer on the version convincingly to put the claims of elegance before those of of verbal correspondence (to put decor before utilitas, to use the terms of the 1689 preface). Not only does he say, as others had said bfefore, that it is an error to make the construction of the Latin follow the Greek, but he also says that is is preferable often to use circumlocution rather than to render a word with another word that does not give the exact sense. (A good example might be the rendering incipientes for e7iap^d|xevoi discussed above.) Nor does he recommend servile adherence to the scholia in difficult passages. He would not have been in favour of the wordy glosses sometimes incorporated in the revisions of 1570 and of Aemilius Portus. He specifically advises that between two Latin expressions rendering the sense of the Greek the most noble and poetic should be chosen. In this light he does not think it an indispensible rule to render all the Greek particles in Latin. If every xe, apa and ye is rendered, then the result in Latin is, of course, something ponderous and barbarous. These principles are put into action in specific comments on the opening eight lines of the version of 1689. He recommends translating ^ifiviv . . . o\)>x)|ievr|v as iram gravem et exitiosam because fif|viv does not signify an ordinary anger. Here oijA,o(ievTiv is not rendered by one word (pemiciosam in 1689) but by a circumlocution which is not only explanatory but gives emphasis to what is a crucial thematic point. In the second verse, \xupV . . aX,Yea ought to be translated as dolores mille rather than sexcentos dolores (1689), a phrase he finds low and prosaic. In verse eight, he proposes to change Quisnam eos Deorum contentione commisit ut pugnarentl to Ecquis

N. Hepp, Homere en France au XVII' siecle (Paris 1968) 568-71. 1 96 Illinois Classical Studies 21(1 996) deorum eos commisit ut per contentionem pugnarentl Presumably this change is in the interests of more idiomatic Latin. Finally, he notes that there are three occurrences of autem in verses 3, 4 and 5, which should be reduced at least to one. However, no subsequent editor took the trouble to revise the versio according to these principles. The consequences of the failure to produce a better and less repellant versio can easily be imagined. Indeed the fears of Casaubon for its effect on tyrones are unlikely to be exaggerated. The reaction against Homer in La Querelle des Anciens et des Modernes in the late seventeenth century in France may in part be an honest response to Homer known principally through the medium of the versio Latina. Boileau in his Reflexions sur Long in of 1694 exposes the tactic of Charles Perrault, who knew little Greek, in translating bad Latin into low French in his Paralleles des Anciens et des Modernes, in which Homer is castigated amongst other things for his ignorance and lack of la politesse.^^ In the preface to his earlier Traite du Sublime of 1674, Boileau had made the acute observation that bad translations into Latin differ in the pemiciousness of their effect from vernacular translations (where the reader may have a better understanding that he is not receiving the original meaning):

II est aise a un Traducteur Latin de se tuer d' affaire aux endroits meme

qu'il n'entend pas. II n'a qu'a traduire le Grec mot pour mot, et a debiter des paroles qu'on peut au moins soupfonner d'etre intelligibles. En effet, le Lecteur, qui bien souvent n'y con^oit rien, s'en prend plutot a soi-meme

qu'a I'ignorance du Traducteur. II n'en est pas ainsi des traductions en langue vulgaire. Tout ce que le Lecteur n'entend point s'appelle un galimatias, dont le Traducteur tout seul est responsible.^'

It is easy for a Latin translator to kill meaning in those particular parts that he does not understand. He has only to translate the Greek word for word and to make those words that he can at least guess comprehensible. In effect the reader, who very often knows nothing, takes on himself the ignorance of the translator. The case is different with translations into the vernacular. All that the reader does not understand at all can be called a

hotchpotch for which the translator is wholly responsible.

Paradoxically, the ad verbum method could be said to have encouraged and perpetuated nonsense in that a careless or ignorant interpreter (Leontius being the obvious if extreme example) might only feel obliged to give approximate equivalents to each word without also feeling the need to string the words together to make overall sense. We may recall here the verdict of Petrarch's biographer Pierre de Nolhac on Leontius: "On comprend que si

^^ Boileau. Oeuvres completes, introduction par A. Adam, textes etablis et annotes par F. Escal (Bruges 1966) "Reflexions critiques sur quelques passages du rheteur Longin," Reflexion in, pp. 504-05. 1

Robin Sowerby 197

Pilate a adopte le systeme de traduction 'verbum ad verbum,' c'est qu'il ^^ etait trop ignorant pour en pouvoir un autre." The versio Latina, therefore was both the product and cause of inattention and ignorance. The difficulties involved were real, but not so great after all the ancillary material had been printed as they had been in the early Renaissance. It is tempting to conclude that the poor state of the versio must be a general reflection of the parlous state of Greek studies more generally, yet Casaubon specifically states that Homer's fate is particularly bad—other Greek authors had fared better. Perhaps these are the philosophers or historians, for they cannot have been what are now considered to be the classics of Greek literature in drama and lyric, in which there was even less interest than in Homer to judge from the number and quality of the editions. However the poor state of the versio is certainly a reflection of the poor state of Homeric studies and while the failure to produce a better one was often a failure of knowledge and scholarship it was also a failure of conviction and will. If Homer had seriously been felt to be the Prince of Poets as he is often called on the title pages of editions, then his text and its interpretation would have called forth the kind of sustained scholarly enterprise and attention accorded to the New Testament or to Virgil. m Appendix: Ilias Graecolatina 1-16

1537: Andreas Divus 1551: Vuolphius 1570: Anon, apud Crispinum 1580: Franciscus Portus 1583: Spondanus 1589: Stephanus 1609: Aemilius Portus 1656: Schrevelius 1689: Anon., Cantabrigiae

Line 1: Mf|viv cxei6e, 0ed, nT|A,r|id5eco 'AxiXfioq

1537 Iram cane Dea Pelidae AchilUs 1551 Iram die Dea Pelidae Achillis 1570 Iram cane dea (musa) Pelidae Achillis 1580 Iram cane dea Pelidae Achillis 1583 Iram cane dea Pelidae Achillis 1589 Iram cane Dea Pelidae Achillis 1609 Iram cane dea Pelidae Achillis

^^ P. de Nolhac, "Les scholies inddites de Petrarque sur Homere," Revue de Philologie 1 (1887) 105. 1 ,

198 Illinois Classical Studies 21 (1996)

1 656 Iram cane Dea (musa) Pelidae Achillis 1689 Iram cane Dea Pelidae Achillis

Line 2: o\)A,OM,evT|v, r\ jx-upC 'Axaioiq akye' e0T|Kev,

1537 Pemiciosam: quae infinites Achivis dolores inflixit 155 Perdentem, quae infinites Graecis dolores fecit 1570 Pemiciosam, quae infinites Achivis dolores imposuit, 1580 Pestiferam, quae plurimos Achivis dolores imposuit 1583 Pemiciosam, quae infinites Achivis dolores fecit 1589 Pemiciosam, quae infinites Grecis dolores attulit, 1609 Pemiciosam, quae infinites Graecis dolores attulit, 1656 Pemiciosam, quae infinites Achivis dolores inflixit: 1689 Pemiciosam, quae sexcentos Achivis dolores fecit:

6' Line 3: noXXaq i(pOi)io\)^ V'^X^'i "Ai8i Tipoiayev

1537 Multas autem fortes animas inferis misit 155 Multas et generesas animas orco demisit 1570 Multas autem fortes animas Pluteni praemisit 1580 Multas autem fortissimas animas Orco demisit 1583 Multas autem fortes animas Orco ante tempus demisit, 1589 Multasque fortissimas animas orco demisit, 1609 Multasque fortissimas animas orco ante-iustum-tempus-cum- detrimento demisit 1656 Multas autem fortes animas Pluteni praemisit 1689 Multas autem fortes animas orco praemature misit

Line 4: fipcoajv, auxo-uc; 6e eX-obpia xeuxe Kuveooiv

1537 Hereum, ipses autem laniamenta fecit canibus 155 Hereum, ipsosque lacerationes fabricavit canibus 1570 Hereum: ipses autem laniamenta fecit canibus 1580 Hereum: ipses autem praedam dilaniandam fecit canibus 1589 Hereum, ipses vere praedam dilaniandam fecit canibus 1609 Hereum, ipses praedam dilaniandam fecit canibus 1656 Hereum: ipses autem laniamenta fecit canibus 1689 Hereum, ipses autem praedam discerpendam fecit canibus

Line 5: eicovoioi xe notai (Aioq 5' exeXeiexo pot)X,fi)

1537 Avibus omnibus. levis autem perficiebatur voluntas 155 Camiverisque avibus omnibus levisque perfecta voluntas 1570 Avibus omnibus: lovi autem perficiebatur voluntas v. consilium 1580 Alitibusque omnibus: lovis autem perficiebatur consilium 11

Robin Sowerby 199

1583 Avibusque omnibus: (lovis autem perficiebatur consilium) 1589 Alitibusque omnibus (ac lovis consilium exitum habuit) 1609 Alitibusque omnibus (lovis enim perficiebatur consilium) 1656 Avibusque omnibus, (Jovis autem perficiebatur voluntas) 1689 Alitibusque omnibus: (Jovis autem perficiebatur consilium)

Line 6: e^ ov 5ti toc Ttpwxa SiaaxriTriv epiaavte

1537 Ex quo sane primum divisi sunt contendentes 1551 Ex quo primum dissenserunt contendentes 1570 Ex quo sane primum dissenserunt litigantes 1580 Ex quo primum disiuncti sunt altercantes 1583 Ex quo sane primum disiuncti sunt litigantes 1589 Ex quo primum dissenserunt contendentes 1609 Ex quo primum dissenserunt contendentes 1656 Ex quo sane primum dissenserunt litigantes 1689 Ex quo primum disjuncti sunt orta contentione

Line 7: 'AxpEibr\q xe, ava^ dv6p(ov, Kal bioq 'AxikXe'iiC,

1537 Atridesque Rex virorum, et divus Achilles. 1551 Atridesque rex virorum, et divinus Achilles. • 1570 Atridesque rex virorum, et divus Achilles. 1580 Atridesque rex virorum, et nobilis Achilles. 1583 Atridesque rex virorum, et divus Achilles. 1589 Atridesque rex virorum, et dius Achilles. 1609 Atridesque rex virorum, et dius Achilles. 1656 Atridesque rex virorum, et divus Achilles. 1689 Atridesque rex virorum, et nobilis Achilles.

Line 8: ti<; x' ap acpcoe Gewv epi5i ^DverjKe fxdxeaOai;

1537 Quis nam ipso Deorum contentioni commisit pugnare?

1 55 Quique sane ipsos Deorum contentionem commiserit pugnare. 1570 Quisnam ipsos Deorum contentione commisit, ut pugnarent? 1580 Quis nam eos deorum contentione commisit ut pugnarent?

1583 Quisnam ipsos deorum liti commisit ut pugnarent? 1589 Quisnam ipsos deorum contentione commisit pugnare? 1609 Quisnam ipsos deorum contentione commisit pugnare?

1656 Quisnam ipsos Deorum liti commisit ut pugnarent? 1689 Quisnam eos Deorum contentione commisit ut pugnarent?

Line 9: Arjxoui; Kal Aioq moq. 6 ydp PaaiA-fii xoA-ooBeiq

1537 Latonae, et lovis filius, hie enim regi iratus 155 Latonae et Jovis filius. Ille enim regi iratus 1

200 Illinois Classical Studies 21(1 996)

1570 Latonae et lovis filius hie enim regi iratus 1580 Latonae, et lovis filius is enim regi iratus 1583 Latonae et lovis filius. hie enim regi iratus 1589 Latonae et lovis filius, ille enim regi iratus 1609 Latonae et lovis filius. ille enim regi iratus 1656 Latonae et Jovis filius. hie enim regi iratus 1689 Latonae et Jovis filius. Hie enim regi iratus

Line 10: vovoov dva atpaxov (opae KaKr|v, 6?ieK0VT0 Se Xaoi,

1537 Morbum per exercitum suseitavit malum: peribant vero populi

155 1 Morbum per exereitum immisit malum, peribant vero populi 1570 Morbum per exereitum suseitavit malum (peribant autem populi) 1580 Morbum in exereitum immisit pestilentem: interibant autem populi 1583 Morbum per exereitum suseitavit malum, peribant autem populi 1589 Morbum in exereitum exeitavit malum: peribant vero populi 1609 Morbum in exereitum exeitavit malum: peribant vero populi 1656 Morbum per exereitum suseitavit malum, (peribant autem populi) 1689 Morbum per exereitum exeitavit noxium, (peribant autem populi)

Line 1 1 : ovveKa xov Xpijariv TiT{|j.r|a ' dpTixfipa

1537 Quoniam Chrysem inhonoravit saeerdotem 155 Eo quod Chrysen inhonoravit saeerdotem 1570 Quoniam Chrysen inhonoravit saeerdotem 1580 Quoniam Chrysen eontumelia affeeerat saeerdotem 1583 Quoniam Chrysen eontumelia affeeerat saeerdotem 1589 Eo quod Chrysen dedeeoravit saeerdotem 1609 Quia Chrysen ignominia affeeit saeerdotem 1656 Quoniam Chrysen inhonoravit saeerdotem 1689 Quoniam Chiysen eontumelia affeeerat saeerdotem

Line 12: 'AxpeiSriq. byapr[XdEQoaqbdvr]ac, 'Axaiwv,

1537 Atrides. hie enim venit eeleres ad naves Aehivorum

1551 Atrides. is enim venit eurrentes ad naves Graeeorum 1570 Atrides. hie enim venit eeleres ad naves Aehivorum 1580 Atrides. is enim venerat eeleres ad naves Graeeorum 1583 Atrides. ille enim venit eeleres ad naves Aehivorum 1589 Atrides. Is enim venit eitas ad naves Graeeorum 1609 Atrides. Is enim venit eitas ad naves Graeeorum 11

Robin Sowerby 201

1656 Atrides. hie enim venit celeres ad naves Achivorum 1689 Atrides. hie enim venerat celeres ad naves Achivorum

Line 13: Xvoo^evoq xe Gvyatpa, cpepcov x' ocTiepeiai' cxTcoiva,

1537 Liberaturusque filiam, ferensque infinita dona;

1 55 Liberaturus filiam, ferensque infinita dona 1570 Liberaturusque filiam, ferensque infinita precia liberationis 1580 Redempturus filiam, et ferens immensum pretium 1583 Liberaturusque filiam, ferensque infinita precia liberationis 1589 Redempturus filiam, ferensque praeclara munera 1609 Redempturusque filiam, ferensque praeclara munera 1656 Liberaturusque filiam, ferensque infinita precia liberationis 1689 Redempturusque filiam, ferensque infinitum pretium liberationis

Line 14: axeiifxax' e'xfov ev xep<7iv eicriPoX^OD 'A7i6A,A,cl)vo(;

1537 Coronas habens in manibus longeiacuiantis Apollinis 1551 Coronas habens in manibus eminus ferientis ApoUinis 1570 Coronamque habens in manibus longe iaculantis Apollinis 1580 Coronas habens in manibus longe iaculantis Apollinis 1583 Coronamque habens in manibus longe jaculantis Apollinis 1589 Coronas habens in manibus eminus ferientis Apollinis 1609 Coronas habens in manibus eminus-ferientis Apollinis 1656 Coronamque habens in manibus longe-jaculantis Apollinis 1689 Coronamque habens in manibus longe-jaculantis Apollinis

Line 15: xpi^cjecp ocvoc OKfiTixpo), Kal iXiaaexo Jidvxaq 'Axaiotx;,

1537 Aureo cum sceptro: et obsecrabat omnes Achivos.

1 55 Aureo cum sceptro, et precabatur omnes Graecos. 1570 Aureo cum sceptro: et obsecrabat omnes Achivos. 1580 Aureo cum sceptro: et orabat omnes Achivos 1583 Aureo cum sceptro: et obsecrabat omnes Achivos 1589 Aureo super sceptro: et precabatur omnes Graecos, 1609 Aureo super sceptro: et precabatur omnes Graecos. 1656 Aureo cum sceptro: et obsecrabat omnes Achivos, 1689 Aureo cum sceptro: et supplex oravit omnes Achivos

Line 16: 'Axpei6a 6e jxdXiaxa bxxa, Koa)j.T|xope Xadv.

1537 Atridas autem maxime duos principes populorum. 1551 Atridas vero maxime geminos omamenta populorum. 1570 Atridas autem maxime duos principes populorum, 1580 Atridas in primis, duos principes populorum: 202 Illinois Classical Studies 21(1 996)

1583 Atridas autem maxime, duos principes populorum, 1589 Atridas vero maxime, duos imperatores exercitum, 1609 Atridas vero maxime, duos imperatores populorum, 1656 Atridas autem maxime, duos principes populorum. 1689 Atridas autem imprimis, duos duces populorum:

University of Stirling